Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

AZ91 합금 주물 내 연행 결함에 대한 캐리어 가스의 영향

TianLiabJ.M.T.DaviesaXiangzhenZhuc
aUniversity of Birmingham, Birmingham B15 2TT, United Kingdom
bGrainger and Worrall Ltd, Bridgnorth WV15 5HP, United Kingdom
cBrunel Centre for Advanced Solidification Technology, Brunel University London, Kingston Ln, London, Uxbridge UB8 3PH, United Kingdom

Abstract

An entrainment defect (also known as a double oxide film defect or bifilm) acts a void containing an entrapped gas when submerged into a light-alloy melt, thus reducing the quality and reproducibility of the final castings. Previous publications, carried out with Al-alloy castings, reported that this trapped gas could be subsequently consumed by the reaction with the surrounding melt, thus reducing the void volume and negative effect of entrainment defects. Compared with Al-alloys, the entrapped gas within Mg-alloy might be more efficiently consumed due to the relatively high reactivity of magnesium. However, research into the entrainment defects within Mg alloys has been significantly limited. In the present work, AZ91 alloy castings were produced under different carrier gas atmospheres (i.e., SF6/CO2, SF6/air). The evolution processes of the entrainment defects contained in AZ91 alloy were suggested according to the microstructure inspections and thermodynamic calculations. The defects formed in the different atmospheres have a similar sandwich-like structure, but their oxide films contained different combinations of compounds. The use of carrier gases, which were associated with different entrained-gas consumption rates, affected the reproducibility of AZ91 castings.

연행 결함(이중 산화막 결함 또는 이중막이라고도 함)은 경합금 용융물에 잠길 때 갇힌 가스를 포함하는 공극으로 작용하여 최종 주물의 품질과 재현성을 저하시킵니다. Al-합금 주물을 사용하여 수행된 이전 간행물에서는 이 갇힌 가스가 주변 용융물과의 반응에 의해 후속적으로 소모되어 공극 부피와 연행 결함의 부정적인 영향을 줄일 수 있다고 보고했습니다. Al-합금에 비해 마그네슘의 상대적으로 높은 반응성으로 인해 Mg-합금 내에 포집된 가스가 더 효율적으로 소모될 수 있습니다. 그러나 Mg 합금 내 연행 결함에 대한 연구는 상당히 제한적이었습니다. 현재 작업에서 AZ91 합금 주물은 다양한 캐리어 가스 분위기(즉, SF6/CO2, SF6/공기)에서 생산되었습니다. AZ91 합금에 포함된 연행 결함의 진화 과정은 미세 조직 검사 및 열역학 계산에 따라 제안되었습니다. 서로 다른 분위기에서 형성된 결함은 유사한 샌드위치 구조를 갖지만 산화막에는 서로 다른 화합물 조합이 포함되어 있습니다. 다른 동반 가스 소비율과 관련된 운반 가스의 사용은 AZ91 주물의 재현성에 영향을 미쳤습니다.

Keywords

Magnesium alloy, Casting, Oxide film, Bifilm, Entrainment defect, Reproducibility

1. Introduction

As the lightest structural metal available on Earth, magnesium became one of the most attractive light metals over the last few decades. The magnesium industry has consequently experienced a rapid development in the last 20 years [1,2], indicating a large growth in demand for Mg alloys all over the world. Nowadays, the use of Mg alloys can be found in the fields of automobiles, aerospace, electronics and etc.[3,4]. It has been predicted that the global consumption of Mg metals will further increase in the future, especially in the automotive industry, as the energy efficiency requirement of both traditional and electric vehicles further push manufactures lightweight their design [3,5,6].

The sustained growth in demand for Mg alloys motivated a wide interest in the improvement of the quality and mechanical properties of Mg-alloy castings. During a Mg-alloy casting process, surface turbulence of the melt can lead to the entrapment of a doubled-over surface film containing a small quantity of the surrounding atmosphere, thus forming an entrainment defect (also known as a double oxide film defect or bifilm) [7][8][9][10]. The random size, quantity, orientation, and placement of entrainment defects are widely accepted to be significant factors linked to the variation of casting properties [7]. In addition, Peng et al. [11] found that entrained oxides films in AZ91 alloy melt acted as filters to Al8Mn5 particles, trapping them as they settle. Mackie et al. [12] further suggested that entrained oxide films can act to trawl the intermetallic particles, causing them to cluster and form extremely large defects. The clustering of intermetallic compounds made the entrainment defects more detrimental for the casting properties.

Most of the previous studies regarding entrainment defects were carried out on Al-alloys [7,[13][14][15][16][17][18], and a few potential methods have been suggested for diminishing their negative effect on the quality of Al-alloy castings. Nyahumwa et al.,[16] shows that the void volume within entrainment defects could be reduced by a hot isostatic pressing (HIP) process. Campbell [7] suggested the entrained gas within the defects could be consumed due to reaction with the surrounding melt, which was further verified by Raiszedeh and Griffiths [19].The effect of the entrained gas consumption on the mechanical properties of Al-alloy castings has been investigated by [8,9], suggesting that the consumption of the entrained gas promoted the improvement of the casting reproducibility.

Compared with the investigation concerning the defects within Al-alloys, research into the entrainment defects within Mg-alloys has been significantly limited. The existence of entrainment defects has been demonstrated in Mg-alloy castings [20,21], but their behaviour, evolution, as well as entrained gas consumption are still not clear.

In a Mg-alloy casting process, the melt is usually protected by a cover gas to avoid magnesium ignition. The cavities of sand or investment moulds are accordingly required to be flushed with the cover gas prior to the melt pouring [22]. Therefore, the entrained gas within Mg-alloy castings should contain the cover gas used in the casting process, rather than air only, which may complicate the structure and evolution of the corresponding entrainment defects.

SF6 is a typical cover gas widely used for Mg-alloy casting processes [23][24][25]. Although this cover gas has been restricted to use in European Mg-alloy foundries, a commercial report has pointed out that this cover is still popular in global Mg-alloy industry, especially in the countries which dominated the global Mg-alloy production, such as China, Brazil, India, etc. [26]. In addition, a survey in academic publications also showed that this cover gas was widely used in recent Mg-alloy studies [27]. The protective mechanism of SF6 cover gas (i.e., the reaction between liquid Mg-alloy and SF6 cover gas) has been investigated by several previous researchers, but the formation process of the surface oxide film is still not clearly understood, and even some published results are conflicting with each other. In early 1970s, Fruehling [28] found that the surface film formed under SF6 was MgO mainly with traces of fluorides, and suggested that SF6 was absorbed in the Mg-alloy surface film. Couling [29] further noticed that the absorbed SF6 reacted with the Mg-alloy melt to form MgF2. In last 20 years, different structures of the Mg-alloy surface films have been reported, as detailed below.(1)

Single-layered film. Cashion [30,31] used X-ray Photoelectron Spectroscopy (XPS) and Auger Spectroscopy (AES) to identify the surface film as MgO and MgF2. He also found that composition of the film was constant throughout the thickness and the whole experimental holding time. The film observed by Cashion had a single-layered structure created from a holding time from 10 min to 100 min.(2)

Double-layered film. Aarstad et. al [32] reported a doubled-layered surface oxide film in 2003. They observed several well-distributed MgF2 particles attached to the preliminary MgO film and grew until they covered 25–50% of the total surface area. The inward diffusion of F through the outer MgO film was the driving force for the evolution process. This double-layered structure was also supported by Xiong’s group [25,33] and Shih et al. [34].(3)

Triple-layered film. The triple-layered film and its evolution process were reported in 2002 by Pettersen [35]. Pettersen found that the initial surface film was a MgO phase and then gradually evolved to the stable MgF2 phase by the inward diffusion of F. In the final stage, the film has a triple-layered structure with a thin O-rich interlayer between the thick top and bottom MgF2 layers.(4)

Oxide film consisted of discrete particles. Wang et al [36] stirred the Mg-alloy surface film into the melt under a SF6 cover gas, and then inspect the entrained surface film after the solidification. They found that the entrained surface films were not continues as the protective surface films reported by other researchers but composed of discrete particles. The young oxide film was composed of MgO nano-sized oxide particles, while the old oxide films consist of coarse particles (about 1  µm in average size) on one side that contained fluorides and nitrides.

The oxide films of a Mg-alloy melt surface or an entrained gas are both formed due to the reaction between liquid Mg-alloy and the cover gas, thus the above-mentioned research regarding the Mg-alloy surface film gives valuable insights into the evolution of entrainment defects. The protective mechanism of SF6 cover gas (i.e., formation of a Mg-alloy surface film) therefore indicated a potential complicated evolution process of the corresponding entrainment defects.

However, it should be noted that the formation of a surface film on a Mg-alloy melt is in a different situation to the consumption of an entrained gas that is submerged into the melt. For example, a sufficient amount of cover gas was supported during the surface film formation in the studies previously mentioned, which suppressed the depletion of the cover gas. In contrast, the amount of entrained gas within a Mg-alloy melt is finite, and the entrained gas may become fully depleted. Mirak [37] introduced 3.5%SF6/air bubbles into a pure Mg-alloy melt solidifying in a specially designed permanent mould. It was found that the gas bubbles were entirely consumed, and the corresponding oxide film was a mixture of MgO and MgF2. However, the nucleation sites (such as the MgF2 spots observed by Aarstad [32] and Xiong [25,33]) were not observed. Mirak also speculated that the MgF2 formed prior to MgO in the oxide film based on the composition analysis, which was opposite to the surface film formation process reported in previous literatures (i.e., MgO formed prior to MgF2). Mirak’s work indicated that the oxide-film formation of an entrained gas may be quite different from that of surface films, but he did not reveal the structure and evolution of the oxide films.

In addition, the use of carrier gas in the cover gases also influenced the reaction between the cover gas and the liquid Mg-alloy. SF6/air required a higher content of SF6 than did a SF6/CO2 carrier gas [38], to avoid the ignition of molten magnesium, revealing different gas-consumption rates. Liang et.al [39] suggested that carbon was formed in the surface film when CO2 was used as a carrier gas, which was different from the films formed in SF6/air. An investigation into Mg combustion [40] reported a detection of Mg2C3 in the Mg-alloy sample after burning in CO2, which not only supported Liang’s results, but also indicated a potential formation of Mg carbides in double oxide film defects.

The work reported here is an investigation into the behaviour and evolution of entrainment defects formed in AZ91 Mg-alloy castings, protected by different cover gases (i.e., SF6/air and SF6/CO2). These carrier gases have different protectability for liquid Mg alloy, which may be therefore associated with different consumption rates and evolution processes of the corresponding entrained gases. The effect of the entrained-gas consumption on the reproducibility of AZ91 castings was also studied.

2. Experiment

2.1. Melting and casting

Three kilograms AZ91 alloy was melted in a mild steel crucible at 700 ± 5 °C. The composition of the AZ91 alloy has been shown in Table 1. Prior to heating, all oxide scale on the ingot surface was removed by machining. The cover gases used were 0.5%SF6/air or 0.5%SF6/CO2 (vol.%) at a flow rate of 6 L/min for different castings. The melt was degassed by argon with a flow rate of 0.3 L/min for 15 min [41,42], and then poured into sand moulds. Prior to pouring, the sand mould cavity was flushed with the cover gas for 20 min [22]. The residual melt (around 1 kg) was solidified in the crucible.

Table 1. Composition (wt.%) of the AZ91 alloy used in this study.

AlZnMnSiFeNiMg
9.40.610.150.020.0050.0017Residual

Fig. 1(a) shows the dimensions of the casting with runners. A top-filling system was deliberately used to generate entrainment defects in the final castings. Green and Campbell [7,43] suggested that a top-filling system caused more entrainment events (i.e., bifilms) during a casting process, compared with a bottom-filling system. A melt flow simulation (Flow-3D software) of this mould, using Reilly’s model [44] regarding the entrainment events, also predicted that a large amount of bifilms would be contained in the final casting (denoted by the black particles in Fig. 1b).

Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

Shrinkage defects also affect the mechanical properties and reproducibility of castings. Since this study focused on the effect of bifilms on the casting quality, the mould has been deliberately designed to avoid generating shrinkage defects. A solidification simulation using ProCAST software showed that no shrinkage defect would be contained in the final casting, as shown in Fig. 1c. The casting soundness has also been confirmed using a real time X-ray prior to the test bar machining.

The sand moulds were made from resin-bonded silica sand, containing 1wt. % PEPSET 5230 resin and 1wt. % PEPSET 5112 catalyst. The sand also contained 2 wt.% Na2SiF6 to act as an inhibitor [45]. The pouring temperature was 700 ± 5 °C. After the solidification, a section of the runner bars was sent to the Sci-Lab Analytical Ltd for a H-content analysis (LECO analysis), and all the H-content measurements were carried out on the 5th day after the casting process. Each of the castings was machined into 40 test bars for a tensile strength test, using a Zwick 1484 tensile test machine with a clip extensometer. The fracture surfaces of the broken test bars were examined using Scanning Electron Microscope (SEM, Philips JEOL7000) with an accelerating voltage of 5–15 kV. The fractured test bars, residual Mg-alloy solidified in the crucible, and the casting runners were then sectioned, polished and also inspected using the same SEM. The cross-section of the oxide film found on the test-bar fracture surface was exposed by the Focused Ion Beam milling technique (FIB), using a CFEI Quanta 3D FEG FIB-SEM. The oxide film required to be analysed was coated with a platinum layer. Then, a gallium ion beam, accelerated to 30 kV, milled the material substrate surrounding the platinum coated area to expose the cross section of the oxide film. EDS analysis of the oxide film’s cross section was carried out using the FIB equipment at accelerating voltage of 30 kV.

2.2. Oxidation cell

As previously mentioned, several past researchers investigated the protective film formed on a Mg-alloy melt surface [38,39,[46][47][48][49][50][51][52]. During these experiments, the amount of cover gas used was sufficient, thus suppressing the depletion of fluorides in the cover gas. The experiment described in this section used a sealed oxidation cell, which limited the supply of cover gas, to study the evolution of the oxide films of entrainment defects. The cover gas contained in the oxidation cell was regarded as large-size “entrained bubble”.

As shown in Fig. 2, the main body of the oxidation cell was a closed-end mild steel tube which had an inner length of 400 mm, and an inner diameter of 32 mm. A water-cooled copper tube was wrapped around the upper section of the cell. When the tube was heated, the cooling system created a temperature difference between the upper and lower sections, causing the interior gas to convect within the tube. The temperature was monitored by a type-K thermocouple located at the top of the crucible. Nie et al. [53] suggested that the SF6 cover gas would react with the steel wall of the holding furnace when they investigated the surface film of a Mg-alloy melt. To avoid this reaction, the interior surface of the steel oxidation cell (shown in Fig. 2) and the upper half section of the thermocouple were coated with boron nitride (the Mg-alloy was not in contact with boron nitride).

Fig. 2. Schematic of the oxidation cell used to study the evolution of the oxide films of the entrainment defects (unit mm).

During the experiment, a block of solid AZ91 alloy was placed in a magnesia crucible located at the bottom of the oxidation cell. The cell was heated to 100 °C in an electric resistance furnace under a gas flow rate of 1 L/min. The cell was held at this temperature for 20 min, to replace the original trapped atmosphere (i.e. air). Then, the oxidation cell was further heated to 700 °C, melting the AZ91 sample. The gas inlet and exit valves were then closed, creating a sealed environment for oxidation under a limited supply of cover gas. The oxidation cell was then held at 700 ± 10 °C for periods of time from 5 min to 30 min in 5-min intervals. At the end of each holding time, the cell was quenched in water. After cooling to room temperature, the oxidised sample was sectioned, polished, and subsequently examined by SEM.

3. Results

3.1. Structure and composition of the entrainment defects formed in SF6/air

The structure and composition of the entrainment defect formed in the AZ91 castings under a cover gas of 0.5%SF6/air was observed by SEM and EDS. The results indicate that there exist two types of entrainment defects which are sketched in Fig. 3: (1) Type A defect whose oxide film has a traditional single-layered structure and (2) Type B defect, whose oxide film has two layers. The details of these defects were introduced in the following. Here it should be noticed that, as the entrainment defects are also known as biofilms or double oxide film, the oxide films of Type B defect were referred to as “multi-layered oxide film” or “multi-layered structure” in the present work to avoid a confusing description such as “the double-layered oxide film of a double oxide film defect”.

Fig. 3. Schematic of the different types of entrainment defects found in AZ91 castings. (a) Type A defect with a single-layered oxide film and (b) Type B defect with two-layered oxide film.

Fig. 4(a-b) shows a Type A defect having a compact single-layered oxide film with about 0.4 µm thickness. Oxygen, fluorine, magnesium and aluminium were detected in this film (Fig. 4c). It is speculated that oxide film is the mixture of fluoride and oxide of magnesium and aluminium. The detection of fluorine revealed that an entrained cover gas was contained in the formation of this defect. That is to say that the pores shown in Fig. 4(a) were not shrinkage defects or hydrogen porosity, but entrainment defects. The detection of aluminium was different with Xiong and Wang’s previous study [47,48], which showed that no aluminium was contained in their surface film of an AZ91 melt protected by a SF6 cover gas. Sulphur could not be clearly recognized in the element map, but there was a S-peak in the corresponding ESD spectrum.

Fig. 4. (a) A Type A entrainment defect formed in SF6/air and having a single-layered oxide film, (b) the oxide film of this defect, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area highlighted in (b).

Fig. 5(a-b) shows a Type B entrainment defect having a multi-layered oxide film. The compact outer layers of the oxide films were enriched with fluorine and oxygen (Fig. 5c), while their relatively porous inner layers were only enriched with oxygen (i.e., poor in fluorine) and partly grew together, thus forming a sandwich-like structure. Therefore, it is speculated that the outer layer is the mixture of fluoride and oxide, while the inner layer is mainly oxide. Sulphur could only be recognized in the EDX spectrum and could not be clearly identified in the element map, which might be due to the small S-content in the cover gas (i.e., 0.5% volume content of SF6 in the cover gas). In this oxide film, aluminium was contained in the outer layer of this oxide film but could not be clearly detected in the inner layer. Moreover, the distribution of Al seems to be uneven. It can be found that, in the right side of the defect, aluminium exists in the film but its concentration can not be identified to be higher than the matrix. However, there is a small area with much higher aluminium concentration in the left side of the defect. Such an uneven distribution of aluminium was also observed in other defects (shown in the following), and it is the result of the formation of some oxide particles in or under the film.

Fig. 5. (a) A Type B entrainment defect formed in SF6/air and having a multi-layered oxide film, (b) the oxide films of this defect have grown together, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (b).

Figs. 4 and 5 show cross sectional observations of the entrainment defects formed in the AZ91 alloy sample cast under a cover gas of SF6/air. It is not sufficient to characterize the entrainment defects only by the figures observed from the two-dimensional section. To have a further understanding, the surface of the entrainment defects (i.e. the oxide film) was further studied by observing the fracture surface of the test bars.

Fig. 6(a) shows fracture surfaces of an AZ91 alloy tensile test bar produced in SF6/air. Symmetrical dark regions can be seen on both sides of the fracture surfaces. Fig. 6(b) shows boundaries between the dark and bright regions. The bright region consisted of jagged and broken features, while the surface of the dark region was relatively smooth and flat. In addition, the EDS results (Fig. 6c-d and Table 2) show that fluorine, oxygen, sulphur, and nitrogen were only detected in the dark regions, indicating that the dark regions were surface protective films entrained into the melt. Therefore, it could be suggested that the dark regions were an entrainment defect with consideration of their symmetrical nature. Similar defects on fracture surfaces of Al-alloy castings have been previously reported [7]Nitrides were only found in the oxide films on the test-bar fracture surfaces but never detected in the cross-sectional samples shown in Figs. 4 and 5. An underlying reason is that the nitrides contained in these samples may have hydrolysed during the sample polishing process [54].

Fig. 6. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar produced under a cover gas of SF6/air. The dimension of the fracture surface is 5 mm × 6 mm, (b) a section of the boundary between the dark and bright regions shown in (a), (c-d) EDS spectrum of the (c) bright regions and (d) dark regions, (e) schematic of an entrainment defect contained in a test bar.

Table 2. EDS results (wt.%) corresponding to the regions shown in Fig. 6 (cover gas: SF6/air).

Empty CellCOMgFAlZnSN
Dark region in Fig. 6(b)3.481.3279.130.4713.630.570.080.73
Bright region in Fig. 6(b)3.5884.4811.250.68

In conjunction with the cross-sectional observation of the defects shown in Figs. 4 and 5, the structure of an entrainment defect contained in a tensile test bar was sketched as shown in Fig. 6(e). The defect contained an entrained gas enclosed by its oxide film, creating a void section inside the test bar. When the tensile force applied on the defect during the fracture process, the crack was initiated at the void section and propagated along the entrainment defect, since cracks would be propagated along the weakest path [55]. Therefore, when the test bar was finally fractured, the oxide films of entrainment defect appeared on both fracture surfaces of the test bar, as shown in Fig. 6(a).

3.2. Structure and composition of the entrainment defects formed in SF6/CO2

Similar to the entrainment defect formed in SF6/air, the defects formed under a cover gas of 0.5%SF6/CO2 also had two types of oxide films (i.e., single-layered and multi-layered types). Fig. 7(a) shows an example of the entrainment defects containing a multi-layered oxide film. A magnified observation to the defect (Fig. 7b) shows that the inner layers of the oxide films had grown together, presenting a sandwich-like structure, which was similar to the defects formed in an atmosphere of SF6/air (Fig. 5b). An EDS spectrum (Fig. 7c) revealed that the joint area (inner layer) of this sandwich-like structure mainly contained magnesium oxides. Peaks of fluorine, sulphur, and aluminium were recognized in this EDS spectrum, but their amount was relatively small. In contrast, the outer layers of the oxide films were compact and composed of a mixture of fluorides and oxides (Fig. 7d-e).

Fig. 7. (a) An example of entrainment defects formed in SF6/CO2 and having a multi-layered oxide film, (b) magnified observation of the defect, showing the inner layer of the oxide films has grown together, (c) EDS spectrum of the point denoted in (b), (d) outer layer of the oxide film, (e) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (d).

Fig. 8(a) shows an entrainment defect on the fracture surfaces of an AZ91 alloy tensile test bar, which was produced in an atmosphere of 0.5%SF6/CO2. The corresponding EDS results (Table 3) showed that oxide film contained fluorides and oxides. Sulphur and nitrogen were not detected. Besides, a magnified observation (Fig. 8b) indicated spots on the oxide film surface. The diameter of the spots ranged from hundreds of nanometres to a few micron meters.

Fig. 8. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar, produced in an atmosphere of SF6/CO2. The dimension of the fracture surface is 5 mm × 6 mm, (b) surface appearance of the oxide films on the fracture surfaces, showing spots on the film surface.

To further reveal the structure and composition of the oxide film clearly, the cross-section of the oxide film on a test-bar fracture surface was onsite exposed using the FIB technique (Fig. 9). As shown in Fig. 9a, a continuous oxide film was found between the platinum coating layer and the Mg-Al alloy substrate. Fig. 9 (b-c) shows a magnified observation to oxide films, indicating a multi-layered structure (denoted by the red box in Fig. 9c). The bottom layer was enriched with fluorine and oxygen and should be the mixture of fluoride and oxide, which was similar to the “outer layer” shown in Figs. 5 and 7, while the only-oxygen-enriched top layer was similar to the “inner layer” shown in Figs. 5 and 7.

Fig. 9. (a) A cross-sectional observation of the oxide film on the fracture surface of the AZ91 casting produced in SF6/CO2, exposed by FIB, (b) a magnified observation of area highlighted in (a), and (c) SEM-EDS elements map of the area shown in (b), obtained by CFEI Quanta 3D FEG FIB-SEM.

Except the continuous film, some individual particles were also observed in or below the continuous film, as shown in Fig. 9. An Al-enriched particle was detected in the left side of the oxide film shown in Fig. 9b and might be speculated to be spinel Mg2AlO4 because it also contains abundant magnesium and oxygen elements. The existing of such Mg2AlO4 particles is responsible for the high concentration of aluminium in small areas of the observed film and the uneven distribution of aluminium, as shown in Fig. 5(c). Here it should be emphasized that, although the other part of the bottom layer of the continuous oxide film contains less aluminium than this Al-enriched particle, the Fig. 9c indicated that the amount of aluminium in this bottom layer was still non-negligible, especially when comparing with the outer layer of the film. Below the right side of the oxide film shown in Fig. 9b, a particle was detected and speculated to be MgO because it is rich in Mg and O. According to Wang’s result [56], lots of discrete MgO particles can be formed on the surface of the Mg melt by the oxidation of Mg melt and Mg vapor. The MgO particles observed in our present work may be formed due to the same reasons. While, due to the differences in experimental conditions, less Mg melt can be vapored or react with O2, thus only a few of MgO particles formed in our work. An enrichment of carbon was also found in the film, revealing that CO2 was able to react with the melt, thus forming carbon or carbides. This carbon concentration was consistent with the relatively high carbon content of the oxide film shown in Table 3 (i.e., the dark region). In the area next to the oxide film.

Table 3. EDS results (wt.%) corresponding to the regions shown in Fig. 8 (cover gas: SF6/ CO2).

Empty CellCOMgFAlZnSN
Dark region in Fig. 8(a)7.253.6469.823.827.030.86
Bright region in Fig. 8(a)2.100.4482.8313.261.36

This cross-sectional observation of the oxide film on a test bar fracture surface (Fig. 9) further verified the schematic of the entrainment defect shown in Fig. 6(e). The entrainment defects formed in different atmospheres of SF6/CO2 and SF6/air had similar structures, but their compositions were different.

3.3. Evolution of the oxide films in the oxidation cell

The results in Section 3.1 and 3.2 have shown the structures and compositions of entrainment defects formed in AZ91 castings under cover gases of SF6/air and SF6/CO2. Different stages of the oxidation reaction may lead to the different structures and compositions of entrainment defects. Although Campbell has conjectured that an entrained gas may react with the surrounding melt, it is rarely reported that the reaction occurring between the Mg-alloy melt and entrapped cover gas. Previous researchers normally focus on the reaction between a Mg-alloy melt and the cover gas in an open environment [38,39,[46][47][48][49][50][51][52], which was different from the situation of a cover gas trapped into the melt. To further understand the formation of the entrainment defect in an AZ91 alloy, the evolution process of oxide films of the entrainment defect was further studied using an oxidation cell.

Fig. 10 (a and d) shows a surface film held for 5 min in the oxidation cell, protected by 0.5%SF6/air. There was only one single layer consisting of fluoride and oxide (MgF2 and MgO). In this surface film. Sulphur was detected in the EDS spectrum, but its amount was too small to be recognized in the element map. The structure and composition of this oxide film was similar to the single-layered films of entrainment defects shown in Fig. 4.

Fig. 10. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/air and held at 700 °C for (a) 5 min; (b) 10 min; (c) 30 min, and (d-f) the SEM-EDS element maps (using Philips JEOL7000) corresponding to the oxide film shown in (a-c) respectively, (d) 5 min; (e) 10 min; (f) 30 min. The red points in (c and f) are the location references, denoting the boundary of the F-enriched layer in different element maps.

After a holding time of 10 min, a thin (O, S)-enriched top layer (around 700 nm) appeared upon the preliminary F-enriched film, forming a multi-layered structure, as shown in Fig. 10(b and e). The thickness of the (O, S)-enriched top layer increased with increased holding time. As shown in Fig. 10(c and f), the oxide film held for 30 min also had a multi-layered structure, but the thickness of its (O, S)-enriched top layer (around 2.5 µm) was higher than the that of the 10-min oxide film. The multi-layered oxide films shown in Fig. 10(b-c) presented a similar appearance to the films of the sandwich-like defect shown in Fig. 5.

The different structures of the oxide films shown in Fig. 10 indicated that fluorides in the cover gas would be preferentially consumed due to the reaction with the AZ91 alloy melt. After the depletion of fluorides, the residual cover gas reacted further with the liquid AZ91 alloy, forming the top (O, S)-enriched layer in the oxide film. Therefore, the different structures and compositions of entrainment defects shown in Figs. 4 and 5 may be due to an ongoing oxidation reaction between melt and entrapped cover gas.

This multi-layered structure has not been reported in previous publications concerning the protective surface film formed on a Mg-alloy melt [38,[46][47][48][49][50][51]. This may be due to the fact that previous researchers carried out their experiments with an un-limited amount of cover gas, creating a situation where the fluorides in the cover gas were not able to become depleted. Therefore, the oxide film of an entrainment defect had behaviour traits similar to the oxide films shown in Fig. 10, but different from the oxide films formed on the Mg-alloy melt surface reported in [38,[46][47][48][49][50][51].

Similar with the oxide films held in SF6/air, the oxide films formed in SF6/CO2 also had different structures with different holding times in the oxidation cell. Fig. 11(a) shows an oxide film, held on an AZ91 melt surface under a cover gas of 0.5%SF6/CO2 for 5 min. This film had a single-layered structure consisting of MgF2. The existence of MgO could not be confirmed in this film. After the holding time of 30 min, the film had a multi-layered structure; the inner layer was of a compact and uniform appearance and composed of MgF2, while the outer layer is the mixture of MgF2 and MgO. Sulphur was not detected in this film, which was different from the surface film formed in 0.5%SF6/air. Therefore, fluorides in the cover gas of 0.5%SF6/CO2 were also preferentially consumed at an early stage of the film growth process. Compared with the film formed in SF6/air, the MgO in film formed in SF6/CO2 appeared later and sulphide did not appear within 30 min. It may mean that the formation and evolution of film in SF6/air is faster than SF6/CO2. CO2 may have subsequently reacted with the melt to form MgO, while sulphur-containing compounds accumulated in the cover gas and reacted to form sulphide in very late stage (may after 30 min in oxidation cell).

Fig. 11. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/CO2, and their SEM-EDS element maps (using Philips JEOL7000). They were held at 700 °C for (a) 5 min; (b) 30 min. The red points in (b) are the location references, denoting the boundary between the top and bottom layers in the oxide film.

4. Discussion

4.1. Evolution of entrainment defects formed in SF6/air

HSC software from Outokumpu HSC Chemistry for Windows (http://www.hsc-chemistry.net/) was used to carry out thermodynamic calculations needed to explore the reactions which might occur between the trapped gases and liquid AZ91 alloy. The solutions to the calculations suggest which products are most likely to form in the reaction process between a small amount of cover gas (i.e., the amount within a trapped bubble) and the AZ91-alloy melt.

In the trials, the pressure was set to 1 atm, and the temperature set to 700 °C. The amount of the cover gas was assumed to be 7 × 10−7 kg, with a volume of approximately 0.57 cm3 (3.14 × 10−8 kmol) for 0.5%SF6/air, and 0.35 cm3 (3.12 × 10−8 kmol) for 0.5%SF6/CO2. The amount of the AZ91 alloy melt in contact with the trapped gas was assumed to be sufficient to complete all reactions. The decomposition products of SF6 were SF5, SF4, SF3, SF2, F2, S(g), S2(g) and F(g) [57][58][59][60].

Fig. 12 shows the equilibrium diagram of the thermodynamic calculation of the reaction between the AZ91 alloy and 0.5%SF6/air. In the diagram, the reactants and products with less than 10−15 kmol have not been shown, as this was 5 orders of magnitude less than the amount of SF6 present (≈ 1.57 × 10−10 kmol) and therefore would not affect the observed process in a practical way.

Fig. 12. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/air and a sufficient amount of AZ91 alloy. The X axis is the amount of AZ91 alloy melt having reacted with the entrained gas, and the vertical Y-axis is the amount of the reactants and products.

This reaction process could be divided into 3 stages.

Stage 1: The formation of fluorides. the AZ91 melt preferentially reacted with SF6 and its decomposition products, producing MgF2, AlF3, and ZnF2. However, the amount of ZnF2 may have been too small to be detected practically (1.25 × 10−12 kmol of ZnF2 compared with 3 × 10−10 kmol of MgF2), which may be the reason why Zn was not detected in any the oxide films shown in Sections 3.13.3. Meanwhile, sulphur accumulated in the residual gas as SO2.

Stage 2: The formation of oxides. After the liquid AZ91 alloy had depleted all the available fluorides in the entrapped gas, the amount of AlF3 and ZnF2 quickly reduced due to a reaction with Mg. O2(g) and SO2 reacted with the AZ91 melt, forming MgO, Al2O3, MgAl2O4, ZnO, ZnSO4 and MgSO4. However, the amount of ZnO and ZnSO4 would have been too small to be found practically by EDS (e.g. 9.5 × 10−12 kmol of ZnO,1.38 × 10−14 kmol of ZnSO4, in contrast to 4.68 × 10−10 kmol of MgF2, when the amount of AZ91 on the X-axis is 2.5 × 10−9 kmol). In the experimental cases, the concentration of F in the cover gas is very low, whole the concentration f O is much higher. Therefore, the stage 1 and 2, i.e, the formation of fluoride and oxide may happen simultaneously at the beginning of the reaction, resulting in the formation of a singer-layered mixture of fluoride and oxide, as shown in Figs. 4 and 10(a). While an inner layer consisted of oxides but fluorides could form after the complete depletion of F element in the cover gas.

Stages 1- 2 theoretically verified the formation process of the multi-layered structure shown in Fig. 10.

The amount of MgAl2O4 and Al2O3 in the oxide film was of a sufficient amount to be detected, which was consistent with the oxide films shown in Fig. 4. However, the existence of aluminium could not be recognized in the oxide films grown in the oxidation cell, as shown in Fig. 10. This absence of Al may be due to the following reactions between the surface film and AZ91 alloy melt:(1)

Al2O3 + 3Mg + = 3MgO + 2Al, △G(700 °C) = -119.82 kJ/mol(2)

Mg + MgAl2O4 = MgO + Al, △G(700 °C) =-106.34 kJ/molwhich could not be simulated by the HSC software since the thermodynamic calculation was carried out under an assumption that the reactants were in full contact with each other. However, in a practical process, the AZ91 melt and the cover gas would not be able to be in contact with each other completely, due to the existence of the protective surface film.

Stage 3: The formation of Sulphide and nitride. After a holding time of 30 min, the gas-phase fluorides and oxides in the oxidation cell had become depleted, allowing the melt reaction with the residual gas, forming an additional sulphur-enriched layer upon the initial F-enriched or (F, O)-enriched surface film, thus resulting in the observed multi-layered structure shown in Fig. 10 (b and c). Besides, nitrogen reacted with the AZ91 melt until all reactions were completed. The oxide film shown in Fig. 6 may correspond to this reaction stage due to its nitride content. However, the results shows that the nitrides were not detected in the polished samples shown in Figs. 4 and 5, but only found on the test bar fracture surfaces. The nitrides may have hydrolysed during the sample preparation process, as follows [54]:(3)

Mg3N2 + 6H2O =3Mg(OH)2 + 2NH3↑(4)

AlN+ 3H2O =Al(OH)3 + NH3

In addition, Schmidt et al. [61] found that Mg3N2 and AlN could react to form ternary nitrides (Mg3AlnNn+2, n= 1, 2, 3…). HSC software did not contain the database of ternary nitrides, and it could not be added into the calculation. The oxide films in this stage may also contain ternary nitrides.

4.2. Evolution of entrainment defects formed in SF6/CO2

Fig. 13 shows the results of the thermodynamic calculation between AZ91 alloy and 0.5%SF6/CO2. This reaction processes can also be divided into three stages.

Fig. 13. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/CO2 and a sufficient amount of AZ91 alloy. The X axis denotes the amount of Mg alloy melt having reacted with the entrained gas, and the vertical Y-axis denotes the amounts of the reactants and products.

Stage 1: The formation of fluorides. SF6 and its decomposition products were consumed by the AZ91 melt, forming MgF2, AlF3, and ZnF2. As in the reaction of AZ91 in 0.5%SF6/air, the amount of ZnF2 was too small to be detected practically (1.51 × 10−13 kmol of ZnF2 compared with 2.67 × 10−10 kmol of MgF2). Sulphur accumulated in the residual trapped gas as S2(g) and a portion of the S2(g) reacted with CO2, to form SO2 and CO. The products in this reaction stage were consistent with the film shown in Fig. 11(a), which had a single layer structure that contained fluorides only.

Stage 2: The formation of oxides. AlF3 and ZnF2 reacted with the Mg in the AZ91 melt, forming MgF2, Al and Zn. The SO2 began to be consumed, producing oxides in the surface film and S2(g) in the cover gas. Meanwhile, the CO2 directly reacted with the AZ91 melt, forming CO, MgO, ZnO, and Al2O3. The oxide films shown in Figs. 9 and 11(b) may correspond to this reaction stage due to their oxygen-enriched layer and multi-layered structure.

The CO in the cover gas could further react with the AZ91 melt, producing C. This carbon may further react with Mg to form Mg carbides, when the temperature reduced (during solidification period) [62]. This may be the reason for the high carbon content in the oxide film shown in Figs. 89. Liang et al. [39] also reported carbon-detection in an AZ91 alloy surface film protected by SO2/CO2. The produced Al2O3 may be further combined with MgO, forming MgAl2O4 [63]. As discussed in Section 4.1, the alumina and spinel can react with Mg, causing an absence of aluminium in the surface films, as shown in Fig. 11.

Stage 3: The formation of Sulphide. the AZ91 melt began to consume S2(g) in the residual entrapped gas, forming ZnS and MgS. These reactions did not occur until the last stage of the reaction process, which could be the reason why the S-content in the defect shown Fig. 7(c) was small.

In summary, thermodynamic calculations indicate that the AZ91 melt will react with the cover gas to form fluorides firstly, then oxides and sulphides in the last. The oxide film in the different reaction stages would have different structures and compositions.

4.3. Effect of the carrier gases on consumption of the entrained gas and the reproducibility of AZ91 castings

The evolution processes of entrainment defects, formed in SF6/air and SF6/CO2, have been suggested in Sections 4.1 and 4.2. The theoretical calculations were verified with respect to the corresponding oxide films found in practical samples. The atmosphere within an entrainment defect could be efficiently consumed due to the reaction with liquid Mg-alloy, in a scenario dissimilar to the Al-alloy system (i.e., nitrogen in an entrained air bubble would not efficiently react with Al-alloy melt [64,65], however, nitrogen would be more readily consumed in liquid Mg alloys, commonly referred to as “nitrogen burning” [66]).

The reaction between the entrained gas and the surrounding liquid Mg-alloy converted the entrained gas into solid compounds (e.g. MgO) within the oxide film, thus reducing the void volume of the entrainment defect and hence probably causing a collapse of the defect (e.g., if an entrained gas of air was depleted by the surrounding liquid Mg-alloy, under an assumption that the melt temperature is 700 °C and the depth of liquid Mg-alloy is 10 cm, the total volume of the final solid products would be 0.044% of the initial volume taken by the entrapped air).

The relationship between the void volume reduction of entrainment defects and the corresponding casting properties has been widely studied in Al-alloy castings. Nyahumwa and Campbell [16] reported that the Hot Isostatic Pressing (HIP) process caused the entrainment defects in Al-alloy castings to collapse and their oxide surfaces forced into contact. The fatigue lives of their castings were improved after HIP. Nyahumwa and Campbell [16] also suggested a potential bonding of the double oxide films that were in contact with each other, but there was no direct evidence to support this. This binding phenomenon was further investigated by Aryafar et.al.[8], who re-melted two Al-alloy bars with oxide skins in a steel tube and then carried out a tensile strength test on the solidified sample. They found that the oxide skins of the Al-alloy bars strongly bonded with each other and became even stronger with an extension of the melt holding time, indicating a potential “healing” phenomenon due to the consumption of the entrained gas within the double oxide film structure. In addition, Raidszadeh and Griffiths [9,19] successfully reduced the negative effect of entrainment defects on the reproducibility of Al-alloy castings, by extending the melt holding time before solidification, which allowed the entrained gas to have a longer time to react with the surrounding melt.

With consideration of the previous work mentioned, the consumption of the entrained gas in Mg-alloy castings may diminish the negative effect of entrainment defects in the following two ways.

(1) Bonding phenomenon of the double oxide films. The sandwich-like structure shown in Fig. 5 and 7 indicated a potential bonding of the double oxide film structure. However, more evidence is required to quantify the increase in strength due to the bonding of the oxide films.

(2) Void volume reduction of entrainment defects. The positive effect of void-volume reduction on the quality of castings has been widely demonstrated by the HIP process [67]. As the evolution processes discussed in Section 4.14.2, the oxide films of entrainment defects can grow together due to an ongoing reaction between the entrained gas and surrounding AZ91 alloy melt. The volume of the final solid products was significant small compared with the entrained gas (i.e., 0.044% as previously mentioned).

Therefore, the consumption rate of the entrained gas (i.e., the growth rate of oxide films) may be a critical parameter for improving the quality of AZ91 alloy castings. The oxide film growth rate in the oxidization cell was accordingly further investigated.

Fig. 14 shows a comparison of the surface film growth rates in different cover gases (i.e., 0.5%SF6/air and 0.5%SF6/CO2). 15 random points on each sample were selected for film thickness measurements. The 95% confidence interval (95%CI) was computed under an assumption that the variation of the film thickness followed a Gaussian distribution. It can be seen that all the surface films formed in 0.5%SF6/air grew faster than those formed in 0.5%SF6/CO2. The different growth rates suggested that the entrained-gas consumption rate of 0.5%SF6/air was higher than that of 0.5%SF6/CO2, which was more beneficial for the consumption of the entrained gas.

Fig. 14. A comparison of the AZ91 alloy oxide film growth rates in 0.5%SF6/air and 0.5%SF6/CO2

It should be noted that, in the oxidation cell, the contact area of liquid AZ91 alloy and cover gas (i.e. the size of the crucible) was relatively small with consideration of the large volume of melt and gas. Consequently, the holding time for the oxide film growth within the oxidation cell was comparatively long (i.e., 5–30 min). However, the entrainment defects contained in a real casting are comparatively very small (i.e., a few microns size as shown in Figs. 36, and [7]), and the entrained gas is fully enclosed by the surrounding melt, creating a relatively large contact area. Hence the reaction time for cover gas and the AZ91 alloy melt may be comparatively short. In addition, the solidification time of real Mg-alloy sand castings can be a few minutes (e.g. Guo [68] reported that a Mg-alloy sand casting with 60 mm diameter required 4 min to be solidified). Therefore, it can be expected that an entrained gas trapped during an Mg-alloy melt pouring process will be readily consumed by the surrounding melt, especially for sand castings and large-size castings, where solidification times are long.

Therefore, the different cover gases (0.5%SF6/air and 0.5%SF6/CO2) associated with different consumption rates of the entrained gases may affect the reproducibility of the final castings. To verify this assumption, the AZ91 castings produced in 0.5%SF6/air and 0.5%SF6/CO2 were machined into test bars for mechanical evaluation. A Weibull analysis was carried out using both linear least square (LLS) method and non-linear least square (non-LLS) method [69].

Fig. 15(a-b) shows a traditional 2-p linearized Weibull plot of the UTS and elongation of the AZ91 alloy castings, obtained by the LLS method. The estimator used is P= (i-0.5)/N, which was suggested to cause the lowest bias among all the popular estimators [69,70]. The casting produced in SF6/air has an UTS Weibull moduli of 16.9, and an elongation Weibull moduli of 5.0. In contrast, the UTS and elongation Weibull modulus of the casting produced in SF6/CO2 are 7.7 and 2.7 respectively, suggesting that the reproducibility of the casting protected by SF6/CO2 were much lower than that produced in SF6/air.

Fig. 15. The Weibull modulus of AZ91 castings produced in different atmospheres, estimated by (a-b) the linear least square method, (c-d) the non-linear least square method, where SSR is the sum of residual squares.

In addition, the author’s previous publication [69] demonstrated a shortcoming of the linearized Weibull plots, which may cause a higher bias and incorrect R2 interruption of the Weibull estimation. A Non-LLS Weibull estimation was therefore carried out, as shown in Fig. 15 (c-d). The UTS Weibull modulus of the SF6/air casting was 20.8, while the casting produced under SF6/CO2 had a lower UTS Weibull modulus of 11.4, showing a clear difference in their reproducibility. In addition, the SF6/air elongation (El%) dataset also had a Weibull modulus (shape = 5.8) higher than the elongation dataset of SF6/CO2 (shape = 3.1). Therefore, both the LLS and Non-LLS estimations suggested that the SF6/air casting has a higher reproducibility than the SF6/CO2 casting. It supports the method that the use of air instead of CO2 contributes to a quicker consumption of the entrained gas, which may reduce the void volume within the defects. Therefore, the use of 0.5%SF6/air instead of 0.5%SF6/CO2 (which increased the consumption rate of the entrained gas) improved the reproducibility of the AZ91 castings.

However, it should be noted that not all the Mg-alloy foundries followed the casting process used in present work. The Mg-alloy melt in present work was degassed, thus reducing the effect of hydrogen on the consumption of the entrained gas (i.e., hydrogen could diffuse into the entrained gas, potentially suppressing the depletion of the entrained gas [7,71,72]). In contrast, in Mg-alloy foundries, the Mg-alloy melt is not normally degassed, since it was widely believed that there is not a ‘gas problem’ when casting magnesium and hence no significant change in tensile properties [73]. Although studies have shown the negative effect of hydrogen on the mechanical properties of Mg-alloy castings [41,42,73], a degassing process is still not very popular in Mg-alloy foundries.

Moreover, in present work, the sand mould cavity was flushed with the SF6 cover gas prior to pouring [22]. However, not all the Mg-alloy foundries flushed the mould cavity in this way. For example, the Stone Foundry Ltd (UK) used sulphur powder instead of the cover-gas flushing. The entrained gas within their castings may be SO2/air, rather than the protective gas.

Therefore, although the results in present work have shown that using air instead of CO2 improved the reproducibility of the final casting, it still requires further investigations to confirm the effect of carrier gases with respect to different industrial Mg-alloy casting processes.

7. Conclusion

Entrainment defects formed in an AZ91 alloy were observed. Their oxide films had two types of structure: single-layered and multi-layered. The multi-layered oxide film can grow together forming a sandwich-like structure in the final casting.2.

Both the experimental results and the theoretical thermodynamic calculations demonstrated that fluorides in the trapped gas were depleted prior to the consumption of sulphur. A three-stage evolution process of the double oxide film defects has been suggested. The oxide films contained different combinations of compounds, depending on the evolution stage. The defects formed in SF6/air had a similar structure to those formed in SF6/CO2, but the compositions of their oxide films were different. The oxide-film formation and evolution process of the entrainment defects were different from that of the Mg-alloy surface films previous reported (i.e., MgO formed prior to MgF2).3.

The growth rate of the oxide film was demonstrated to be greater under SF6/air than SF6/CO2, contributing to a quicker consumption of the damaging entrapped gas. The reproducibility of an AZ91 alloy casting improved when using SF6/air instead of SF6/CO2.

Acknowledgements

The authors acknowledge funding from the EPSRC LiME grant EP/H026177/1, and the help from Dr W.D. Griffiths and Mr. Adrian Carden (University of Birmingham). The casting work was carried out in University of Birmingham.

Reference

[1]

M.K. McNutt, SALAZAR K.

Magnesium, Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

Reston, Virginia (2013)

Google Scholar[2]

Magnesium

Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

(1996)

Google Scholar[3]

I. Ostrovsky, Y. Henn

ASTEC’07 International Conference-New Challenges in Aeronautics, Moscow (2007), pp. 1-5

Aug 19-22

View Record in ScopusGoogle Scholar[4]

Y. Wan, B. Tang, Y. Gao, L. Tang, G. Sha, B. Zhang, N. Liang, C. Liu, S. Jiang, Z. Chen, X. Guo, Y. Zhao

Acta Mater., 200 (2020), pp. 274-286

ArticleDownload PDFView Record in Scopus[5]

J.T.J. Burd, E.A. Moore, H. Ezzat, R. Kirchain, R. Roth

Appl. Energy, 283 (2021), Article 116269

ArticleDownload PDFView Record in Scopus[6]

A.M. Lewis, J.C. Kelly, G.A. Keoleian

Appl. Energy, 126 (2014), pp. 13-20

ArticleDownload PDFView Record in Scopus[7]

J. Campbell

Castings

Butterworth-Heinemann, Oxford (2004)

Google Scholar[8]

M. Aryafar, R. Raiszadeh, A. Shalbafzadeh

J. Mater. Sci., 45 (2010), pp. 3041-3051 View PDF

CrossRefView Record in Scopus[9]

R. Raiszadeh, W.D. Griffiths

Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 42 (2011), pp. 133-143 View PDF

CrossRefView Record in Scopus[10]

R. Raiszadeh, W.D. Griffiths

J. Alloy. Compd., 491 (2010), pp. 575-580

ArticleDownload PDFView Record in Scopus[11]

L. Peng, G. Zeng, T.C. Su, H. Yasuda, K. Nogita, C.M. Gourlay

JOM, 71 (2019), pp. 2235-2244 View PDF

CrossRefView Record in Scopus[12]

S. Ganguly, A.K. Mondal, S. Sarkar, A. Basu, S. Kumar, C. Blawert

Corros. Sci., 166 (2020)[13]

G.E. Bozchaloei, N. Varahram, P. Davami, S.K. Kim

Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 548 (2012), pp. 99-105

View Record in Scopus[14]

S. Fox, J. Campbell

Scr. Mater., 43 (2000), pp. 881-886

ArticleDownload PDFView Record in Scopus[15]

M. Cox, R.A. Harding, J. Campbell

Mater. Sci. Technol., 19 (2003), pp. 613-625

View Record in Scopus[16]

C. Nyahumwa, N.R. Green, J. Campbell

Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 32 (2001), pp. 349-358

View Record in Scopus[17]

A. Ardekhani, R. Raiszadeh

J. Mater. Eng. Perform., 21 (2012), pp. 1352-1362 View PDF

CrossRefView Record in Scopus[18]

X. Dai, X. Yang, J. Campbell, J. Wood

Mater. Sci. Technol., 20 (2004), pp. 505-513

View Record in Scopus[19]

E.M. Elgallad, M.F. Ibrahim, H.W. Doty, F.H. Samuel

Philos. Mag., 98 (2018), pp. 1337-1359 View PDF

CrossRefView Record in Scopus[20]

W.D. Griffiths, N.W. Lai

Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 190-196 View PDF

CrossRefView Record in Scopus[21]

A.R. Mirak, M. Divandari, S.M.A. Boutorabi, J. Campbell

Int. J. Cast Met. Res., 20 (2007), pp. 215-220 View PDF

CrossRefView Record in Scopus[22]

C. Cingi

Laboratory of Foundry Engineering

Helsinki University of Technology, Espoo, Finland (2006)

Google Scholar[23]

Y. Jia, J. Hou, H. Wang, Q. Le, Q. Lan, X. Chen, L. Bao

J. Mater. Process. Technol., 278 (2020), Article 116542

ArticleDownload PDFView Record in Scopus[24]

S. Ouyang, G. Yang, H. Qin, S. Luo, L. Xiao, W. Jie

Mater. Sci. Eng. A, 780 (2020), Article 139138

ArticleDownload PDFView Record in Scopus[25]

S.-m. Xiong, X.-F. Wang

Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

ArticleDownload PDFView Record in Scopus[26]

G.V. Research

Grand View Research

(2018)

USA

Google Scholar[27]

T. Li, J. Davies

Metall. Mater. Trans. A, 51 (2020), pp. 5389-5400 View PDF

CrossRefView Record in Scopus[28]J.F. Fruehling, The University of Michigan, 1970.

Google Scholar[29]

S. Couling

36th Annual World Conference on Magnesium, Norway (1979), pp. 54-57

View Record in ScopusGoogle Scholar[30]

S. Cashion, N. Ricketts, P. Hayes

J. Light Met., 2 (2002), pp. 43-47

ArticleDownload PDFView Record in Scopus[31]

S. Cashion, N. Ricketts, P. Hayes

J. Light Met., 2 (2002), pp. 37-42

ArticleDownload PDFView Record in Scopus[32]

K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

Various Techniques to Study the Surface of Magnesium Protected by SF6

TMS (2003)

Google Scholar[33]

S.-M. Xiong, X.-L. Liu

Metall. Mater. Trans. A, 38 (2007), pp. 428-434 View PDF

CrossRefView Record in Scopus[34]

T.-S. Shih, J.-B. Liu, P.-S. Wei

Mater. Chem. Phys., 104 (2007), pp. 497-504

ArticleDownload PDFView Record in Scopus[35]

G. Pettersen, E. Øvrelid, G. Tranell, J. Fenstad, H. Gjestland

Mater. Sci. Eng. A, 332 (2002), pp. 285-294

ArticleDownload PDFView Record in Scopus[36]

H. Bo, L.B. Liu, Z.P. Jin

J. Alloy. Compd., 490 (2010), pp. 318-325

ArticleDownload PDFView Record in Scopus[37]

A. Mirak, C. Davidson, J. Taylor

Corros. Sci., 52 (2010), pp. 1992-2000

ArticleDownload PDFView Record in Scopus[38]

B.D. Lee, U.H. Beak, K.W. Lee, G.S. Han, J.W. Han

Mater. Trans., 54 (2013), pp. 66-73 View PDF

View Record in Scopus[39]

W.Z. Liang, Q. Gao, F. Chen, H.H. Liu, Z.H. Zhao

China Foundry, 9 (2012), pp. 226-230 View PDF

CrossRef[40]

U.I. Gol’dshleger, E.Y. Shafirovich

Combust. Explos. Shock Waves, 35 (1999), pp. 637-644[41]

A. Elsayed, S.L. Sin, E. Vandersluis, J. Hill, S. Ahmad, C. Ravindran, S. Amer Foundry

Trans. Am. Foundry Soc., 120 (2012), pp. 423-429[42]

E. Zhang, G.J. Wang, Z.C. Hu

Mater. Sci. Technol., 26 (2010), pp. 1253-1258

View Record in Scopus[43]

N.R. Green, J. Campbell

Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 173 (1993), pp. 261-266

ArticleDownload PDFView Record in Scopus[44]

C Reilly, MR Jolly, NR Green

Proceedings of MCWASP XII – 12th Modelling of Casting, Welding and Advanced Solidifcation Processes, Vancouver, Canada (2009)

Google Scholar[45]H.E. Friedrich, B.L. Mordike, Springer, Germany, 2006.

Google Scholar[46]

C. Zheng, B.R. Qin, X.B. Lou

Proceedings of the 2010 International Conference on Mechanical, Industrial, and Manufacturing Technologies, ASME (2010), pp. 383-388

Mimt 2010 View PDF

CrossRefView Record in ScopusGoogle Scholar[47]

S.M. Xiong, X.F. Wang

Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

ArticleDownload PDFView Record in Scopus[48]

S.M. Xiong, X.L. Liu

Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 428-434 View PDF

CrossRefView Record in Scopus[49]

T.S. Shih, J.B. Liu, P.S. Wei

Mater. Chem. Phys., 104 (2007), pp. 497-504

ArticleDownload PDFView Record in Scopus[50]

K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

Magn. Technol. (2003), pp. 5-10[51]

G. Pettersen, E. Ovrelid, G. Tranell, J. Fenstad, H. Gjestland

Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 332 (2002), pp. 285-294

ArticleDownload PDFView Record in Scopus[52]

X.F. Wang, S.M. Xiong

Corros. Sci., 66 (2013), pp. 300-307

ArticleDownload PDFView Record in Scopus[53]

S.H. Nie, S.M. Xiong, B.C. Liu

Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 422 (2006), pp. 346-351

ArticleDownload PDFView Record in Scopus[54]

C. Bauer, A. Mogessie, U. Galovsky

Zeitschrift Fur Metallkunde, 97 (2006), pp. 164-168 View PDF

CrossRef[55]

Q.G. Wang, D. Apelian, D.A. Lados

J. Light Met., 1 (2001), pp. 73-84

ArticleDownload PDFView Record in Scopus[56]

S. Wang, Y. Wang, Q. Ramasse, Z. Fan

Metall. Mater. Trans. A, 51 (2020), pp. 2957-2974[57]

S. Hayashi, W. Minami, T. Oguchi, H.J. Kim

Kag. Kog. Ronbunshu, 35 (2009), pp. 411-415 View PDF

CrossRefView Record in Scopus[58]

K. Aarstad

Norwegian University of Science and Technology

(2004)

Google Scholar[59]

R.L. Wilkins

J. Chem. Phys., 51 (1969), p. 853

-&

View Record in Scopus[60]

O. Kubaschewski, K. Hesselemam

Thermo-Chemical Properties of Inorganic Substances

Springer-Verlag, Belin (1991)

Google Scholar[61]

R. Schmidt, M. Strobele, K. Eichele, H.J. Meyer

Eur. J. Inorg. Chem. (2017), pp. 2727-2735 View PDF

CrossRefView Record in Scopus[62]

B. Hu, Y. Du, H. Xu, W. Sun, W.W. Zhang, D. Zhao

J. Min. Metall. Sect. B-Metall., 46 (2010), pp. 97-103

View Record in Scopus[63]

O. Salas, H. Ni, V. Jayaram, K.C. Vlach, C.G. Levi, R. Mehrabian

J. Mater. Res., 6 (1991), pp. 1964-1981

View Record in Scopus[64]

S.S.S. Kumari, U.T.S. Pillai, B.C. Pai

J. Alloy. Compd., 509 (2011), pp. 2503-2509

ArticleDownload PDFView Record in Scopus[65]

H. Scholz, P. Greil

J. Mater. Sci., 26 (1991), pp. 669-677

View Record in Scopus[66]

P. Biedenkopf, A. Karger, M. Laukotter, W. Schneider

Magn. Technol., 2005 (2005), pp. 39-42

View Record in Scopus[67]

H.V. Atkinson, S. Davies

Metall. Mater. Trans. A, 31 (2000), pp. 2981-3000 View PDF

CrossRefView Record in Scopus[68]

E.J. Guo, L. Wang, Y.C. Feng, L.P. Wang, Y.H. Chen

J. Therm. Anal. Calorim., 135 (2019), pp. 2001-2008 View PDF

CrossRefView Record in Scopus[69]

T. Li, W.D. Griffiths, J. Chen

Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 48A (2017), pp. 5516-5528 View PDF

CrossRefView Record in Scopus[70]

M. Tiryakioglu, D. Hudak

J. Mater. Sci., 42 (2007), pp. 10173-10179 View PDF

CrossRefView Record in Scopus[71]

Y. Yue, W.D. Griffiths, J.L. Fife, N.R. Green

Proceedings of the 1st International Conference on 3d Materials Science (2012), pp. 131-136 View PDF

CrossRefView Record in ScopusGoogle Scholar[72]

R. Raiszadeh, W.D. Griffiths

Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 37 (2006), pp. 865-871

View Record in Scopus[73]

Z.C. Hu, E.L. Zhang, S.Y. Zeng

Mater. Sci. Technol., 24 (2008), pp. 1304-1308 View PDF

CrossRefView Record in Scopus

Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

플라즈마 회전 전극 공정 중 분말 형성에 대한 공정 매개변수 및 냉각 가스의 영향

Effects of process parameters and cooling gas on powder formation during the plasma rotating electrode process

Yujie Cuia Yufan Zhaoa1 Haruko Numatab Kenta Yamanakaa Huakang Biana Kenta Aoyagia AkihikoChibaa
aInstitute for Materials Research, Tohoku University, Sendai 980-8577, JapanbDepartment of Materials Processing, Graduate School of Engineering, Tohoku University, Sendai 980-8577, Japan

Highlights

•The limitation of increasing the rotational speed in decreasing powder size was clarified.

•Cooling and disturbance effects varied with the gas flowing rate.

•Inclined angle of the residual electrode end face affected powder formation.

•Additional cooling gas flowing could be applied to control powder size.

Abstract

The plasma rotating electrode process (PREP) is rapidly becoming an important powder fabrication method in additive manufacturing. However, the low production rate of fine PREP powder limits the development of PREP. Herein, we investigated different factors affecting powder formation during PREP by combining experimental methods and numerical simulations. The limitation of increasing the rotation electrode speed in decreasing powder size is attributed to the increased probability of adjacent droplets recombining and the decreased tendency of granulation. The effects of additional Ar/He gas flowing on the rotational electrode on powder formation is determined through the cooling effect, the disturbance effect, and the inclined effect of the residual electrode end face simultaneously. A smaller-sized powder was obtained in the He atmosphere owing to the larger inclined angle of the residual electrode end face compared to the Ar atmosphere. Our research highlights the route for the fabrication of smaller-sized powders using PREP.

플라즈마 회전 전극 공정(PREP)은 적층 제조 에서 중요한 분말 제조 방법으로 빠르게 자리잡고 있습니다. 그러나 미세한 PREP 분말의 낮은 생산율은 PREP의 개발을 제한합니다. 여기에서 우리는 실험 방법과 수치 시뮬레이션을 결합하여 PREP 동안 분말 형성에 영향을 미치는 다양한 요인을 조사했습니다. 분말 크기 감소에서 회전 전극 속도 증가의 한계는 인접한 액적 재결합 확률 증가 및 과립화 경향 감소에 기인합니다.. 회전 전극에 흐르는 추가 Ar/He 가스가 분말 형성에 미치는 영향은 냉각 효과, 외란 효과 및 잔류 전극 단면의 경사 효과를 통해 동시에 결정됩니다. He 분위기에서는 Ar 분위기에 비해 잔류 전극 단면의 경사각이 크기 때문에 더 작은 크기의 분말이 얻어졌다. 우리의 연구는 PREP를 사용하여 더 작은 크기의 분말을 제조하는 경로를 강조합니다.

Keywords

Plasma rotating electrode process

Ti-6Al-4 V alloy, Rotating speed, Numerical simulation, Gas flowing, Powder size

Introduction

With the development of additive manufacturing, there has been a significant increase in high-quality powder production demand [1,2]. The initial powder characteristics are closely related to the uniform powder spreading [3,4], packing density [5], and layer thickness observed during additive manufacturing [6], thus determining the mechanical properties of the additive manufactured parts [7,8]. Gas atomization (GA) [9–11], centrifugal atomization (CA) [12–15], and the plasma rotating electrode process (PREP) are three important powder fabrication methods.

Currently, GA is the dominant powder fabrication method used in additive manufacturing [16] for the fabrication of a wide range of alloys [11]. GA produces powders by impinging a liquid metal stream to droplets through a high-speed gas flow of nitrogen, argon, or helium. With relatively low energy consumption and a high fraction of fine powders, GA has become the most popular powder manufacturing technology for AM.

The entrapped gas pores are generally formed in the powder after solidification during GA, in which the molten metal is impacted by a high-speed atomization gas jet. In addition, satellites are formed in GA powder when fine particles adhere to partially molten particles.

The gas pores of GA powder result in porosity generation in the additive manufactured parts, which in turn deteriorates its mechanical properties because pores can become crack initiation sites [17]. In CA, a molten metal stream is poured directly onto an atomizer disc spinning at a high rotational speed. A thin film is formed on the surface of the disc, which breaks into small droplets due to the centrifugal force. Metal powder is obtained when these droplets solidify.

Compared with GA powder, CA powder exhibits higher sphericity, lower impurity content, fewer satellites, and narrower particle size distribution [12]. However, very high speed is required to obtain fine powder by CA. In PREP, the molten metal, melted using the plasma arc, is ejected from the rotating rod through centrifugal force. Compared with GA powder, PREP-produced powders also have higher sphericity and fewer pores and satellites [18].

For instance, PREP-fabricated Ti6Al-4 V alloy powder with a powder size below 150 μm exhibits lower porosity than gas-atomized powder [19], which decreases the porosity of additive manufactured parts. Furthermore, the process window during electron beam melting was broadened using PREP powder compared to GA powder in Inconel 718 alloy [20] owing to the higher sphericity of the PREP powder.

In summary, PREP powder exhibits many advantages and is highly recommended for powder-based additive manufacturing and direct energy deposition-type additive manufacturing. However, the low production rate of fine PREP powder limits the widespread application of PREP powder in additive manufacturing.

Although increasing the rotating speed is an effective method to decrease the powder size [21,22], the reduction in powder size becomes smaller with the increased rotating speed [23]. The occurrence of limiting effects has not been fully clarified yet.

Moreover, the powder size can be decreased by increasing the rotating electrode diameter [24]. However, these methods are quite demanding for the PREP equipment. For instance, it is costly to revise the PREP equipment to meet the demand of further increasing the rotating speed or electrode diameter.

Accordingly, more feasible methods should be developed to further decrease the PREP powder size. Another factor that influences powder formation is the melting rate [25]. It has been reported that increasing the melting rate decreases the powder size of Inconel 718 alloy [26].

In contrast, the powder size of SUS316 alloy was decreased by decreasing the plasma current within certain ranges. This was ascribed to the formation of larger-sized droplets from fluid strips with increased thickness and spatial density at higher plasma currents [27]. The powder size of NiTi alloy also decreases at lower melting rates [28]. Consequently, altering the melting rate, varied with the plasma current, is expected to regulate the PREP powder size.

Furthermore, gas flowing has a significant influence on powder formation [27,29–31]. On one hand, the disturbance effect of gas flowing promotes fluid granulation, which in turn contributes to the formation of smaller-sized powder [27]. On the other hand, the cooling effect of gas flowing facilitates the formation of large-sized powder due to increased viscosity and surface tension. However, there is a lack of systematic research on the effect of different gas flowing on powder formation during PREP.

Herein, the authors systematically studied the effects of rotating speed, electrode diameter, plasma current, and gas flowing on the formation of Ti-6Al-4 V alloy powder during PREP as additive manufactured Ti-6Al-4 V alloy exhibits great application potential [32]. Numerical simulations were conducted to explain why increasing the rotating speed is not effective in decreasing powder size when the rotation speed reaches a certain level. In addition, the different factors incited by the Ar/He gas flowing on powder formation were clarified.

Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.
Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

References

[1] W. Ding, G. Chen, M. Qin, Y. He, X. Qu, Low-cost Ti powders for additive manufacturing treated by fluidized bed, Powder Technol. 350 (2019) 117–122, https://doi.org/
10.1016/j.powtec.2019.03.042.
[2] W.S.W. Harun, M.S.I.N. Kamariah, N. Muhamad, S.A.C. Ghani, F. Ahmad, Z. Mohamed,
A review of powder additive manufacturing processes for metallic biomaterials,
Powder Technol. 327 (2018) 128–151, https://doi.org/10.1016/j.powtec.2017.12.
058.
[3] M. Ahmed, M. Pasha, W. Nan, M. Ghadiri, A simple method for assessing powder
spreadability for additive manufacturing, Powder Technol. 367 (2020) 671–679,
https://doi.org/10.1016/j.powtec.2020.04.033.
[4] W. Nan, M. Pasha, M. Ghadiri, Numerical simulation of particle flow and segregation
during roller spreading process in additive manufacturing, Powder Technol. 364
(2020) 811–821, https://doi.org/10.1016/j.powtec.2019.12.023.
[5] A. Averardi, C. Cola, S.E. Zeltmann, N. Gupta, Effect of particle size distribution on the
packing of powder beds : a critical discussion relevant to additive manufacturing,
Mater. Today Commun. 24 (2020) 100964, https://doi.org/10.1016/j.mtcomm.
2020.100964.
[6] K. Riener, N. Albrecht, S. Ziegelmeier, R. Ramakrishnan, L. Haferkamp, A.B. Spierings,
G.J. Leichtfried, Influence of particle size distribution and morphology on the properties of the powder feedstock as well as of AlSi10Mg parts produced by laser powder bed fusion (LPBF), Addit. Manuf. 34 (2020) 101286, https://doi.org/10.1016/j.
addma.2020.101286.
[7] W.S.W. Harun, N.S. Manam, M.S.I.N. Kamariah, S. Sharif, A.H. Zulkifly, I. Ahmad, H.
Miura, A review of powdered additive manufacturing techniques for Ti-6Al-4V biomedical applications, Powder Technol. 331 (2018) 74–97, https://doi.org/10.1016/j.
powtec.2018.03.010.
[8] A.T. Sutton, C.S. Kriewall, M.C. Leu, J.W. Newkirk, Powder characterisation techniques and effects of powder characteristics on part properties in powder-bed fusion processes, Virtual Phys. Prototyp. 12 (2017) (2017) 3–29, https://doi.org/10.
1080/17452759.2016.1250605.
[9] G. Chen, Q. Zhou, S.Y. Zhao, J.O. Yin, P. Tan, Z.F. Li, Y. Ge, J. Wang, H.P. Tang, A pore
morphological study of gas-atomized Ti-6Al-4V powders by scanning electron microscopy and synchrotron X-ray computed tomography, Powder Technol. 330
(2018) 425–430, https://doi.org/10.1016/j.powtec.2018.02.053.
[10] Y. Feng, T. Qiu, Preparation, characterization and microwave absorbing properties of
FeNi alloy prepared by gas atomization method, J. Alloys Compd. 513 (2012)
455–459, https://doi.org/10.1016/j.jallcom.2011.10.079.

[11] I.E. Anderson, R.L. Terpstra, Progress toward gas atomization processing with increased uniformity and control, Mater. Sci. Eng. A 326 (2002) 101–109, https://
doi.org/10.1016/S0921-5093(01)01427-7.
[12] P. Phairote, T. Plookphol, S. Wisutmethangoon, Design and development of a centrifugal atomizer for producing zinc metal powder, Int. J. Appl. Phys. Math. 2 (2012)
77–82, https://doi.org/10.7763/IJAPM.2012.V2.58.
[13] L. Tian, I. Anderson, T. Riedemann, A. Russell, Production of fine calcium powders by
centrifugal atomization with rotating quench bath, Powder Technol. 308 (2017)
84–93, https://doi.org/10.1016/j.powtec.2016.12.011.
[14] M. Eslamian, J. Rak, N. Ashgriz, Preparation of aluminum/silicon carbide metal matrix composites using centrifugal atomization, Powder Technol. 184 (2008) 11–20,
https://doi.org/10.1016/j.powtec.2007.07.045.
[15] T. Plookphol, S. Wisutmethangoon, S. Gonsrang, Influence of process parameters on
SAC305 lead-free solder powder produced by centrifugal atomization, Powder
Technol. 214 (2011) 506–512, https://doi.org/10.1016/j.powtec.2011.09.015.
[16] M.Z. Gao, B. Ludwig, T.A. Palmer, Impact of atomization gas on characteristics of austenitic stainless steel powder feedstocks for additive manufacturing, Powder
Technol. 383 (2021) 30–42, https://doi.org/10.1016/j.powtec.2020.12.005.
[17] X. Shui, K. Yamanaka, M. Mori, Y. Nagata, A. Chiba, Effects of post-processing on cyclic fatigue response of a titanium alloy additively manufactured by electron beam
melting, Mater. Sci. Eng. A 680 (2017) 239–248, https://doi.org/10.1016/j.msea.
2016.10.059.
[18] C. Wang, X.H. Zhao, Y.C. Ma, Q.X. Wang, Y.J. Lai, S.J. Liang, Study of the spherical
HoCu powders prepared by supreme-speed plasma rotating electrode process,
Powder Metallurgy Technology 38 (3) (2020) 227–233, https://doi.org/10.19591/
j.cnki.cn11-1974/tf.2020.03.011 (in Chinese).
[19] G. Chen, S.Y. Zhao, P. Tan, J. Wang, C.S. Xiang, H.P. Tang, A comparative study of Ti6Al-4V powders for additive manufacturing by gas atomization, plasma rotating
electrode process and plasma atomization, Powder Technol. 333 (2018) 38–46,
https://doi.org/10.1016/j.powtec.2018.04.013.
[20] Y. Zhao, K. Aoyagi, Y. Daino, K. Yamanaka, A. Chiba, Significance of powder feedstock
characteristics in defect suppression of additively manufactured Inconel 718, Addit.
Manuf. 34 (2020) 101277, https://doi.org/10.1016/j.addma.2020.101277.
[21] Y. Nie, J. Tang, B. Yang, Q. Lei, S. Yu, Y. Li, Comparison in characteristic and atomization behavior of metallic powders produced by plasma rotating electrode process,
Adv. Powder Technol. 31 (2020) 2152–2160, https://doi.org/10.1016/j.apt.2020.03.
006.
[22] Y. Cui, Y. Zhao, H. Numata, H. Bian, K. Wako, K. Yamanaka, K. Aoyagi, C. Zhang, A.
Chiba, Effects of plasma rotating electrode process parameters on the particle size
distribution and microstructure of Ti-6Al-4 V alloy powder, Powder Technol 376
(2020) 363–372, https://doi.org/10.1016/j.powtec.2020.08.027.
[23] J. Tang, Y. Nie, Q. Lei, Y. Li, Characteristics and atomization behavior of Ti-6Al-4V
powder produced by plasma rotating electrode process Adv, Powder Technol. 10
(2019) 2330–2337, https://doi.org/10.1016/j.apt.2019.07.015.
[24] M. Zdujić, D. Uskoković, Production of atomized metal and alloy powders by the rotating electrode process, Sov. Powder Metall. Met. Ceram. 29 (1990) 673–683,
https://doi.org/10.1007/BF00795571.
[25] L. Zhang, Y. Zhao, Particle size distribution of tin powder produced by centrifugal
atomisation using rotating cups, Powder Technol. 318 (2017) 62–67, https://doi.
org/10.1016/j.powtec.2017.05.038.
[26] Y. Liu, S. Liang, Z. Han, J. Song, Q. Wang, A novel model of calculating particle sizes in
plasma rotating electrode process for superalloys, Powder Technol. 336 (2018)
406–414, https://doi.org/10.1016/j.powtec.2018.06.002.
[27] Y. Zhao, Y. Cui, H. Numata, H. Bian, K. Wako, K. Yamanaka, Centrifugal granulation
behavior in metallic powder fabrication by plasma rotating electrode process, Sci.
Rep. (2020) 1–15, https://doi.org/10.1038/s41598-020-75503-w.
[28] T. Hsu, C. Wei, L. Wu, Y. Li, A. Chiba, M. Tsai, Nitinol powders generate from plasma
rotation electrode process provide clean powder for biomedical devices used with
suitable size, spheroid surface and pure composition, Sci. Rep. 8 (2018) 1–8,
https://doi.org/10.1038/s41598-018-32101-1.
[29] M. Wei, S. Chen, M. Sun, J. Liang, C. Liu, M. Wang, Atomization simulation and preparation of 24CrNiMoY alloy steel powder using VIGA technology at high gas pressure, Powder Technol. 367 (2020) 724–739, https://doi.org/10.1016/j.powtec.
2020.04.030.
[30] Y. Tan, X. Zhu, X.Y. He, B. Ding, H. Wang, Q. Liao, H. Li, Granulation characteristics of
molten blast furnace slag by hybrid centrifugal-air blast technique, Powder Technol.
323 (2018) 176–185, https://doi.org/10.1016/j.powtec.2017.09.040.
[31] P. Xu, D.H. Liu, J. Hu, G.Y. Lin, Synthesis of Ni-Ti composite powder by radio frequency plasma spheroidization process, Nonferrous Metals Science and Engineering
39 (1) (2020) 67–71 , (in Chinese) 10.13264/j.cnki.ysjskx.2020.01.011.
[32] H. Mehboob, F. Tarlochan, A. Mehboob, S.H. Chang, S. Ramesh, W.S.W. Harun, K.
Kadirgama, A novel design, analysis and 3D printing of Ti-6Al-4V alloy bioinspired porous femoral stem, J. Mater. Sci. Mater. Med. 31 (2020) 78, https://doi.
org/10.1007/s10856-020-06420-7.
[33] FLOW-3D® Version 11.2 [Computer software]. , Flow Science, Inc., Santa Fe, NM,
2017https://www.flow3d.com.
[34] M. Boivineau, C. Cagran, D. Doytier, V. Eyraud, M.H. Nadal, B. Wilthan, G. Pottlacher,
Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy, Int. J.
Thermophys. 27 (2006) 507–529, https://doi.org/10.1007/PL00021868.
[35] J. Liu, Q. Qin, Q. Yu, The effect of size distribution of slag particles obtained in dry
granulation on blast furnace slag cement strength, Powder Technol. 362 (2020)
32–36, https://doi.org/10.1016/j.powtec.2019.11.115.
[36] M. Tanaka, S. Tashiro, A study of thermal pinch effect of welding arcs, J. Japan Weld.
Soc. 25 (2007) 336–342, https://doi.org/10.2207/qjjws.25.336 (in Japanese).
[37] T. Kamiya, A. Kayano, Disintegration of viscous fluid in the ligament state purged
from a rotating disk, J. Chem. Eng. JAPAN. 4 (1971) 364–369, https://doi.org/10.
1252/jcej.4.364.
[38] T. Kamiya, An analysis of the ligament-type disintegration of thin liquid film at the
edge of a rotating disk, J. Chem. Eng. Japan. 5 (1972) 391–396, https://doi.org/10.
1252/jcej.5.391.
[39] J. Burns, C. Ramshaw, R. Jachuck, Measurement of liquid film thickness and the determination of spin-up radius on a rotating disc using an electrical resistance technique, Chem. Eng. Sci. 58 (2003) 2245–2253, https://doi.org/10.1016/S0009-2509
(03)00091-5.
[40] J. Rauscher, R. Kelly, J. Cole, An asymptotic solution for the laminar flow of a thin film
on a rotating disk, J. Appl. Mech. Trans. ASME 40 (1973) 43–47, https://doi.org/10.
1115/1.3422970

Figure 2. Schematic diagram for pilot-scale cooling-water circulation system (a) along with a real picture of the system (b).

Application of Computational Fluid Dynamics in Chlorine-Dynamics Modeling of In-Situ Chlorination Systems for Cooling Systems

Jongchan Yi 1, Jonghun Lee 1, Mohd Amiruddin Fikri 2,3, Byoung-In Sang 4 and Hyunook Kim 1,*

Abstract

염소화는 상대적인 효율성과 저렴한 비용으로 인해 발전소 냉각 시스템에서 생물학적 오염을 제어하는​​데 선호되는 방법입니다. 해안 지역에 발전소가 있는 경우 바닷물을 사용하여 현장에서 염소를 전기화학적으로 생성할 수 있습니다. 이를 현장 전기염소화라고 합니다. 이 접근 방식은 유해한 염소화 부산물이 적고 염소를 저장할 필요가 없다는 점을 포함하여 몇 가지 장점이 있습니다. 그럼에도 불구하고, 이 전기화학적 공정은 실제로는 아직 초기 단계에 있습니다. 이 연구에서는 파일럿 규모 냉각 시스템에서 염소 붕괴를 시뮬레이션하기 위해 병렬 1차 동역학을 적용했습니다. 붕괴가 취수관을 따라 발생하기 때문에 동역학은 전산유체역학(CFD) 코드에 통합되었으며, 이후에 파이프의 염소 거동을 시뮬레이션하는데 적용되었습니다. 실험과 시뮬레이션 데이터는 강한 난류가 형성되는 조건하에서도 파이프 벽을 따라 염소 농도가 점진적인 것으로 나타났습니다. 염소가 중간보다 파이프 표면을 따라 훨씬 더 집중적으로 남아 있다는 사실은 전기 염소화를 기반으로 하는 시스템의 전체 염소 요구량을 감소시킬 수 있었습니다. 현장 전기 염소화 방식의 냉각 시스템은 직접 주입 방식에 필요한 염소 사용량의 1/3만 소비했습니다. 따라서 현장 전기염소화는 해안 지역의 발전소에서 바이오파울링 제어를 위한 비용 효율적이고 환경 친화적인 접근 방식으로 사용될 수 있다고 결론지었습니다.

Chlorination is the preferred method to control biofouling in a power plant cooling system due to its comparative effectiveness and low cost. If a power plant is located in a coastal area, chlorine can be electrochemically generated in-situ using seawater, which is called in-situ electrochlorination; this approach has several advantages including fewer harmful chlorination byproducts and no need for chlorine storage. Nonetheless, this electrochemical process is still in its infancy in practice. In this study, a parallel first-order kinetics was applied to simulate chlorine decay in a pilot-scale cooling system. Since the decay occurs along the water-intake pipe, the kinetics was incorporated into computational fluid dynamics (CFD) codes, which were subsequently applied to simulate chlorine behavior in the pipe. The experiment and the simulation data indicated that chlorine concentrations along the pipe wall were incremental, even under the condition where a strong turbulent flow was formed. The fact that chlorine remained much more concentrated along the pipe surface than in the middle allowed for the reduction of the overall chlorine demand of the system based on the electro-chlorination. The cooling system, with an in-situ electro-chlorination, consumed only 1/3 of the chlorine dose demanded by the direct injection method. Therefore, it was concluded that in-situ electro-chlorination could serve as a cost-effective and environmentally friendly approach for biofouling control at power plants on coastal areas.

Keywords

computational fluid dynamics; power plant; cooling system; electro-chlorination; insitu chlorination

Figure 1. Electrodes and batch experiment set-up. (a) Two cylindrical electrodes used in this study. (b) Batch experiment set-up for kinetic tests.
Figure 1. Electrodes and batch experiment set-up. (a) Two cylindrical electrodes used in this study. (b) Batch experiment set-up for kinetic tests.
Figure 2. Schematic diagram for pilot-scale cooling-water circulation system (a) along with a real picture of the system (b).
Figure 2. Schematic diagram for pilot-scale cooling-water circulation system (a) along with a real picture of the system (b).
Figure 3. Free chlorine decay curves in seawater with different TOC and initial chlorine concentration. Each line represents the predicted concentration of chlorine under a given condition. (a) Artificial seawater solution with 1 mg L−1 of TOC; (b) artificial seawater solution with 2 mg L−1 of TOC; (c) artificial seawater solution with 3 mg L−1 of TOC; (d) West Sea water (1.3 mg L−1 of TOC).
Figure 3. Free chlorine decay curves in seawater with different TOC and initial chlorine concentration. Each line represents the predicted concentration of chlorine under a given condition. (a) Artificial seawater solution with 1 mg L−1 of TOC; (b) artificial seawater solution with 2 mg L−1 of TOC; (c) artificial seawater solution with 3 mg L−1 of TOC; (d) West Sea water (1.3 mg L−1 of TOC).
Figure 4. Correlation between model and experimental data in the chlorine kinetics using seawater.
Figure 4. Correlation between model and experimental data in the chlorine kinetics using seawater.
Figure 5. Free chlorine concentrations in West Sea water under different current conditions in an insitu electro-chlorination system.
Figure 5. Free chlorine concentrations in West Sea water under different current conditions in an insitu electro-chlorination system.
Figure 6. Free chlorine distribution along the sampling ports under different flow rates. Each dot represents experimental data, and each point on the black line is the expected chlorine concentration obtained from computational fluid dynamics (CFD) simulation with a parallel first-order decay model. The red-dotted line is the desirable concentration at the given flow rate: (a) 600 L min−1 of flow rate, (b) 700 L min−1 of flow rate, (c) 800 L min−1 of flow rate, (d) 900 L min−1 of flow rate.
Figure 6. Free chlorine distribution along the sampling ports under different flow rates. Each dot represents experimental data, and each point on the black line is the expected chlorine concentration obtained from computational fluid dynamics (CFD) simulation with a parallel first-order decay model. The red-dotted line is the desirable concentration at the given flow rate: (a) 600 L min−1 of flow rate, (b) 700 L min−1 of flow rate, (c) 800 L min−1 of flow rate, (d) 900 L min−1 of flow rate.
Figure 7. Fluid contour images from CFD simulation of the electro-chlorination experiment. Inlet flow rate is 800 L min−1. Outlet pressure was set to 10.8 kPa. (a) Chlorine concentration; (b) expanded view of electrode side in image (a); (c) velocity magnitude; (d) pressure.
Figure 7. Fluid contour images from CFD simulation of the electro-chlorination experiment. Inlet flow rate is 800 L min−1. Outlet pressure was set to 10.8 kPa. (a) Chlorine concentration; (b) expanded view of electrode side in image (a); (c) velocity magnitude; (d) pressure.
Figure 8. Chlorine concentration contour in the simulation of full-scale in-situ electro-chlorination with different cathode positions. The pipe diameter is 2 m and the flow rate is 14 m3 s−1. The figure shows 10 m of the pipeline. (a) The simulation result when the cathode is placed on the surface of the pipe wall. (b) The simulation result when the cathode is placed on the inside of the pipe with 100 mm of distance from the pipe wall.
Figure 8. Chlorine concentration contour in the simulation of full-scale in-situ electro-chlorination with different cathode positions. The pipe diameter is 2 m and the flow rate is 14 m3 s−1. The figure shows 10 m of the pipeline. (a) The simulation result when the cathode is placed on the surface of the pipe wall. (b) The simulation result when the cathode is placed on the inside of the pipe with 100 mm of distance from the pipe wall.
Figure 9. Comparison of in-situ electro-chlorination and direct chlorine injection in full-scale applications. (a) Estimated chlorine concentrations along the pipe surface. (b) Relative chlorine demands.
Figure 9. Comparison of in-situ electro-chlorination and direct chlorine injection in full-scale applications. (a) Estimated chlorine concentrations along the pipe surface. (b) Relative chlorine demands.

References

  1. Macknick, J.; Newmark, R.; Heath, G.; Hallett, K.C. Operational water consumption and withdrawal factors for electricity generating technologies: A review of existing literature. Environ. Res. Lett. 2012, 7, 045802.
  2. Pan, S.-Y.; Snyder, S.W.; Packman, A.I.; Lin, Y.J.; Chiang, P.-C. Cooling water use in thermoelectric power generation and its associated challenges for addressing water-energy nexus. Water-Energy Nexus 2018, 1, 26–41.
  3. Feeley, T.J., III; Skone, T.J.; Stiegel, G.J., Jr.; McNemar, A.; Nemeth, M.; Schimmoller, B.; Murphy, J.T.;
    Manfredo, L. Water: A critical resource in the thermoelectric power industry. Energy 2008, 33, 1–11.
  4. World Nuclear Association. World Nuclear Performance Report 2016; World Nuclear Association: London, UK, 2016.
  5. Pugh, S.; Hewitt, G.; Müller-Steinhagen, H. Fouling during the use of seawater as coolant—The development of a user guide. Heat Transf. Eng. 2005, 26, 35–43.
  6. Satpathy, K.K.; Mohanty, A.K.; Sahu, G.; Biswas, S.; Prasad, M.; Slvanayagam, M. Biofouling and its control in seawater cooled power plant cooling water system—A review. Nucl. Power 2010, 17, 191–242.
  7. Cristiani, P.; Perboni, G. Antifouling strategies and corrosion control in cooling circuits. Bioelectrochemistry 2014, 97, 120–126.
  8. Walker, M.E.; Safari, I.; Theregowda, R.B.; Hsieh, M.-K.; Abbasian, J.; Arastoopour, H.; Dzombak, D.A.; Miller, D.C. Economic impact of condenser fouling in existing thermoelectric power plants. Energy 2012,44, 429–437.
  9. Yi, J.; Ahn, Y.; Hong, M.; Kim, G.-H.; Shabnam, N.; Jeon, B.; Sang, B.-I.; Kim, H. Comparison between OCl−-Injection and In Situ Electrochlorination in the Formation of Chlorate and Perchlorate in Seawater. Appl.Sci. 2019, 9, 229.
  10. Xue, Y.; Zhao, J.; Qiu, R.; Zheng, J.; Lin, C.; Ma, B.; Wang, P. In Situ glass antifouling using Pt nanoparticle coating for periodic electrolysis of seawater. Appl. Surf. Sci. 2015, 357, 60–68.
  11. Mahfouz, A.B.; Atilhan, S.; Batchelor, B.; Linke, P.; Abdel-Wahab, A.; El-Halwagi, M.M. Optimal scheduling of biocide dosing for seawater-cooled power and desalination plants. Clean Technol. Environ. Policy 2011, 13, 783–796.
  12. Rubio, D.; López-Galindo, C.; Casanueva, J.F.; Nebot, E. Monitoring and assessment of an industrial antifouling treatment. Seasonal effects and influence of water velocity in an open once-through seawater cooling system. Appl. Therm. Eng. 2014, 67, 378–387.
  13. European Integrated Pollution Prevention and Control (IPPC) Bureau, European Commission. Reference Document on the Application of Best Available Techniques to Industrial Cooling Systems December 2001; European Commission, Tech. Rep: Brussels, Belgium, 2001.
  14. Venkatesan R.; Murthy P. S. Macrofouling Control in Power Plants. In Springer Series on Biofilms; Springer: Berlin/Heidelberg, Germany, 2008.
  15. Kastl, G.; Fisher, I.; Jegatheesan, V. Evaluation of chlorine decay kinetics expressions for drinking water distribution systems modelling. J. Water Supply Res. Technol. AQUA 1999, 48, 219–226.
  16. Fisher, I.; Kastl, G.; Sathasivan, A.; Cook, D.; Seneverathne, L. General model of chlorine decay in blends of surface waters, desalinated water, and groundwaters. J. Environ. Eng. 2015, 141, 04015039.
  17. Fisher, I.; Kastl, G.; Sathasivan, A.; Jegatheesan, V. Suitability of chlorine bulk decay models for planning and management of water distribution systems. Crit. Rev. Environ. Sci. Technol. 2011, 41, 1843–1882.
  18. Fisher, I.; Kastl, G.; Sathasivan, A. Evaluation of suitable chlorine bulk-decay models for water distribution systems. Water Res. 2011, 45, 4896–4908.
  19. Haas, C.N.; Karra, S. Kinetics of wastewater chlorine demand exertion. J. (Water Pollut. Control Fed.) 1984, 56, 170–173.
  20. Zeng, J.; Jiang, Z.; Chen, Q.; Zheng, P.; Huang, Y. The decay kinetics of residual chlorine in cooling seawater simulation experiments. Acta Oceanol. Sin. 2009, 28, 54–59.
  21. Saeed, S.; Prakash, S.; Deb, N.; Campbell, R.; Kolluru, V.; Febbo, E.; Dupont, J. Development of a sitespecific kinetic model for chlorine decay and the formation of chlorination by-products in seawater. J. Mar. Sci. Eng. 2015, 3, 772–792.
  22. Al Heboos, S.; Licskó, I. Application and comparison of two chlorine decay models for predicting bulk chlorine residuals. Period. Polytech. Civ. Eng. 2017, 61, 7–13.
  23. Shadloo, M.S.; Oger, G.; Le Touzé, D. Smoothed particle hydrodynamics method for fluid flows, towards industrial applications: Motivations, current state, and challenges. Comput. Fluids 2016, 136, 11–34.
  24. Wols, B.; Hofman, J.; Uijttewaal, W.; Rietveld, L.; Van Dijk, J. Evaluation of different disinfection calculation methods using CFD. Environ. Model. Softw. 2010, 25, 573–582.
  25. Angeloudis, A.; Stoesser, T.; Falconer, R.A. Predicting the disinfection efficiency range in chlorine contact tanks through a CFD-based approach. Water Res. 2014, 60, 118–129.
  26. Zhang, J.; Tejada-Martínez, A.E.; Zhang, Q. Developments in computational fluid dynamics-based modeling for disinfection technologies over the last two decades: A review. Environ. Model. Softw. 2014, 58,71–85.
  27. Lim, Y.H.; Deering, D.D. In Modeling Chlorine Residual in a Ground Water Supply Tank for a Small Community in Cold Conditions, World Environmental and Water Resources Congress 2017; American Society of Civil Engineers: Reston, Virginia, USA, 2017; pp. 124–138.
  28. Hernández-Cervantes, D.; Delgado-Galván, X.; Nava, J.L.; López-Jiménez, P.A.; Rosales, M.; Mora Rodríguez, J. Validation of a computational fluid dynamics model for a novel residence time distribution analysis in mixing at cross-junctions. Water 2018, 10, 733.
  29. Hua, F.; West, J.; Barker, R.; Forster, C. Modelling of chlorine decay in municipal water supplies. Water Res. 1999, 33, 2735–2746.
  30. Jonkergouw, P.M.; Khu, S.-T.; Savic, D.A.; Zhong, D.; Hou, X.Q.; Zhao, H.-B. A variable rate coefficient chlorine decay model. Environ. Sci. Technol. 2009, 43, 408–414.
  31. Nejjari, F.; Puig, V.; Pérez, R.; Quevedo, J.; Cugueró, M.; Sanz, G.; Mirats, J. Chlorine decay model calibration and comparison: Application to a real water network. Procedia Eng. 2014, 70, 1221–1230.
  32. Kohpaei, A.J.; Sathasivan, A.; Aboutalebi, H. Effectiveness of parallel second order model over second and first order models. Desalin. Water Treat. 2011, 32, 107–114.
  33. Powell, J.C.; Hallam, N.B.; West, J.R.; Forster, C.F.; Simms, J. Factors which control bulk chlorine decay rates. Water Res. 2000, 34, 117–126.
  34. Clark, R.M.; Sivaganesan, M. Predicting chlorine residuals in drinking water: Second order model. J. Water Resour. Plan. Manag. 2002, 128, 152–161.
  35. Li, X.; Li, C.; Bayier, M.; Zhao, T.; Zhang, T.; Chen, X.; Mao, X. Desalinated seawater into pilot-scale drinking water distribution system: Chlorine decay and trihalomethanes formation. Desalin. Water Treat. 2016, 57,19149–19159.
  36. United States Environmental Protection Agency (EPA). Chlorine, Total Residual (Spectrophotometric, DPD); EPA-NERL: 330.5; EPA: Cincinnati, OH, USA, 1978.
  37. Polman, H.; Verhaart, F.; Bruijs, M. Impact of biofouling in intake pipes on the hydraulics and efficiency of pumping capacity. Desalin. Water Treat. 2013, 51, 997–1003.
  38. Rajagopal, S.; Van der Velde, G.; Van der Gaag, M.; Jenner, H.A. How effective is intermittent chlorination to control adult mussel fouling in cooling water systems? Water Res. 2003, 37, 329–338.
  39. Bruijs, M.C.; Venhuis, L.P.; Daal, L. Global Experiences in Optimizing Biofouling Control through PulseChlorination®. 2017. Available online: https://www.researchgate.net/publication/318561645_Global_Experiences_in_Optimizing_Biofouling_Co ntrol_through_Pulse-ChlorinationR (accessed on 1 May 2020).
  40. Kim, H.; Hao, O.J.; McAvoy, T.J. Comparison between model-and pH/ORP-based process control for an AAA system. Tamkang J. Sci. Eng. 2000, 3, 165–172.
  41. Brdys, M.; Chang, T.; Duzinkiewicz, K. Intelligent Model Predictive Control of Chlorine Residuals in Water Distribution Systems, Bridging the Gap: Meeting the World’s Water and Environmental Resources Challenges. In Proceedings of the ASCE Water Resource Engineering and Water Resources Planning and Management, July 30–August 2, 2000; pp. 1–11
Fig. 5. The predicted shapes of initial breach (a) Rectangular (b) V-notch. Fig. 6. Dam breaching stages.

Investigating the peak outflow through a spatial embankment dam breach

공간적 제방댐 붕괴를 통한 최대 유출량 조사

Mahmoud T.GhonimMagdy H.MowafyMohamed N.SalemAshrafJatwaryFaculty of Engineering, Zagazig University, Zagazig 44519, Egypt

Abstract

Investigating the breach outflow hydrograph is an essential task to conduct mitigation plans and flood warnings. In the present study, the spatial dam breach is simulated by using a three-dimensional computational fluid dynamics model, FLOW-3D. The model parameters were adjusted by making a comparison with a previous experimental model. The different parameters (initial breach shape, dimensions, location, and dam slopes) are studied to investigate their effects on dam breaching. The results indicate that these parameters have a significant impact. The maximum erosion rate and peak outflow for the rectangular shape are higher than those for the V-notch by 8.85% and 5%, respectively. Increasing breach width or decreasing depth by 5% leads to increasing maximum erosion rate by 11% and 15%, respectively. Increasing the downstream slope angle by 4° leads to an increase in both peak outflow and maximum erosion rate by 2.0% and 6.0%, respectively.

유출 유출 수문곡선을 조사하는 것은 완화 계획 및 홍수 경보를 수행하는 데 필수적인 작업입니다. 본 연구에서는 3차원 전산유체역학 모델인 FLOW-3D를 사용하여 공간 댐 붕괴를 시뮬레이션합니다. 이전 실험 모델과 비교하여 모델 매개변수를 조정했습니다.

다양한 매개변수(초기 붕괴 형태, 치수, 위치 및 댐 경사)가 댐 붕괴에 미치는 영향을 조사하기 위해 연구됩니다. 결과는 이러한 매개변수가 상당한 영향을 미친다는 것을 나타냅니다. 직사각형 형태의 최대 침식율과 최대 유출량은 V-notch보다 각각 8.85%, 5% 높게 나타났습니다.

위반 폭을 늘리거나 깊이를 5% 줄이면 최대 침식률이 각각 11% 및 15% 증가합니다. 하류 경사각을 4° 증가시키면 최대 유출량과 최대 침식률이 각각 2.0% 및 6.0% 증가합니다.

Keywords

Spatial dam breach; FLOW-3D; Overtopping erosion; Computational fluid dynamics (CFD)

1. Introduction

There are many purposes for dam construction, such as protection from flood disasters, water storage, and power generationEmbankment failures may have a catastrophic impact on lives and infrastructure in the downstream regions. One of the most common causes of embankment dam failure is overtopping. Once the overtopping of the dam begins, the breach formation will start in the dam body then end with the dam failure. This failure occurs within a very short time, which threatens to be very dangerous. Therefore, understanding and modeling the embankment breaching processes is essential for conducting mitigation plans, flood warnings, and forecasting flood damage.

The analysis of the dam breaching process is implemented by different techniques: comparative methods, empirical models with dimensional and dimensionless solutions, physical-based models, and parametric models. These models were described in detail [1]Parametric modeling is commonly used to simulate breach growth as a time-dependent linear process and calculate outflow discharge from the breach using hydraulics principles [2]. Alhasan et al. [3] presented a simple one-dimensional mathematical model and a computer code to simulate the dam breaching process. These models were validated by small dams breaching during the floods in 2002 in the Czech Republic. Fread [4] developed an erosion model (BREACH) based on hydraulics principles, sediment transport, and soil mechanics to estimate breach size, time of formation, and outflow discharge. Říha et al. [5] investigated the dam break process for a cascade of small dams using a simple parametric model for piping and overtopping erosion, as well as a 2D shallow-water flow model for the flood in downstream areas. Goodarzi et al. [6] implemented mathematical and statistical methods to assess the effect of inflows and wind speeds on the dam’s overtopping failure.

Dam breaching studies can be divided into two main modes of erosion. The first mode is called “planar dam breach” where the flow overtops the whole dam width. While the second mode is called “spatial dam breach” where the flow overtops through the initial pilot channel (i.e., a channel created in the dam body). Therefore, the erosion will be in both vertical and horizontal directions [7].

The erosion process through the embankment dams occurs due to the shear stress applied by water flows. The dam breaching evolution can be divided into three stages [8][9], but Y. Yang et al. [10] divided the breach development into five stages: Stage I, the seepage erosion; Stage II, the initial breach formation; Stage III, the head erosion; Stage IV, the breach expansion; and Stage V, the re-equilibrium of the river channel through the breach. Many experimental tests have been carried out on non-cohesive embankment dams with an initial breach to examine the effect of upstream inflow discharges on the longitudinal profile evolution and the time to inflection point [11].

Zhang et al. [12] studied the effect of changing downstream slope angle, sediment grain size, and dam crest length on erosion rates. They noticed that increasing dam crest length and decreasing downstream slope angle lead to decreasing sediment transport rate. While the increase in sediment grain size leads to an increased sediment transport rate at the initial stages. Höeg et al. [13] presented a series of field tests to investigate the stability of embankment dams made of various materials. Overtopping and piping were among the failure tests carried out for the dams composed of homogeneous rock-fill, clay, or gravel with a height of up to 6.0 m. Hakimzadeh et al. [14] constructed 40 homogeneous cohesive and non-cohesive embankment dams to study the effect of changing sediment diameter and dam height on the breaching process. They also used genetic programming (GP) to estimate the breach outflow. Refaiy et al. [15] studied different scenarios for the downstream drain geometry, such as length, height, and angle, to minimize the effect of piping phenomena and therefore increase dam safety.

Zhu et al. [16] examined the effect of headcut erosion on dam breach growth, especially in the case of cohesive dams. They found that the breach growth in non-cohesive embankments is slower than cohesive embankments due to the little effect of headcut. Schmocker and Hager [7] proposed a relationship for estimating peak outflow from the dam breach process.(1)QpQin-1=1.7exp-20hc23d5013H0

where: Qp = peak outflow discharge.

Qin = inflow discharge.

hc = critical flow depth.

d50 = mean sediment diameter.

Ho = initial dam height.

Yu et al. [17] carried out an experimental study for homogeneous non-cohesive embankment dams in a 180° bending rectangular flume to determine the effect of overtopping flows on breaching formation. They found that the main factors influencing breach formation are water level, river discharge, and embankment material diameter.

Wu et al. [18] carried out a series of experiments to investigate the effect of breaching geometry on both non-cohesive and cohesive embankment dams in a U-bend flume due to overtopping flows. In the case of non-cohesive embankments, the non-symmetrical lateral expansion was noticed during the breach formation. This expansion was described by a coefficient ranging from 2.7 to 3.3.

The numerical models of the dam breach can be categorized according to different parameters, such as flow dimensions (1D, 2D, or 3D), flow governing equations, and solution methods. The 1D models are mainly used to predict the outflow hydrograph from the dam breach. Saberi et al. [19] applied the 1D Saint-Venant equation, which is solved by the finite difference method to investigate the outflow hydrograph during dam overtopping failure. Because of the ability to study dam profile evolution and breach formation, 2D models are more applicable than 1D models. Guan et al. [20] and Wu et al. [21] employed both 2D shallow water equations (SWEs) and sediment erosion equations, which are solved by the finite volume method to study the effect of the dam’s geometry parameters on outflow hydrograph and dam profile evolution. Wang et al. [22] also proposed a second-order hybrid-type of total variation diminishing (TVD) finite-difference to estimate the breach outflow by solving the 2D (SWEs). The accuracy of (SWEs) for both vertical flow contraction and surface roughness has been assessed [23]. They noted that the accuracy of (SWEs) is acceptable for milder slopes, but in the case of steeper slopes, modelers should be more careful. Generally, the accuracy of 2D models is still low, especially with velocity distribution over the flow depth, lateral momentum exchange, density-driven flows, and bottom friction [24]. Therefore, 3D models are preferred. Larocque et al. [25] and Yang et al. [26] started to use three-dimensional (3D) models that depend on the Reynolds-averaged Navier-Stokes (RANS) equations.

Previous experimental studies concluded that there is no clear relationship between the peak outflow from the dam breach and the initial breach characteristics. Some of these studies depend on the sharp-crested weir fixed at the end of the flume to determine the peak outflow from the breach, which leads to a decrease in the accuracy of outflow calculations at the microscale. The main goals of this study are to carry out a numerical simulation for a spatial dam breach due to overtopping flows by using (FLOW-3D) software to find an empirical equation for the peak outflow discharge from the breach and determine the worst-case that leads to accelerating the dam breaching process.

2. Numerical simulation

The current study for spatial dam breach is simulated by using (FLOW-3D) software [27], which is a powerful computational fluid dynamics (CFD) program.

2.1. Geometric presentations

A stereolithographic (STL) file is prepared for each change in the initial breach geometry and dimensions. The CAD program is useful for creating solid objects and converting them to STL format, as shown in Fig. 1.

2.2. Governing equations

The governing equations for water flow are three-dimensional Reynolds Averaged Navier-Stokes equations (RANS).

The continuity equation:(2)∂ui∂xi=0

The momentum equation:(3)∂ui∂t+1VFuj∂ui∂xj=1ρ∂∂xj-pδij+ν∂ui∂xj+∂uj∂xi-ρu`iu`j¯

where u is time-averaged velocity,ν is kinematic viscosity, VF is fractional volume open to flow, p is averaged pressure and -u`iu`j¯ are components of Reynold’s stress. The Volume of Fluid (VOF) technique is used to simulate the free surface profile. Hirt et al. [28] presented the VOF algorithm, which employs the function (F) to express the occupancy of each grid cell with fluid. The value of (F) varies from zero to unity. Zero value refers to no fluid in the grid cell, while the unity value refers to the grid cell being fully occupied with fluid. The free surface is formed in the grid cells having (F) values between zero and unity.(4)∂F∂t+1VF∂∂xFAxu+∂∂yFAyv+∂∂zFAzw=0

where (u, v, w) are the velocity components in (x, y, z) coordinates, respectively, and (AxAyAz) are the area fractions.

2.3. Boundary and initial conditions

To improve the accuracy of the results, the boundary conditions should be carefully determined. In this study, two mesh blocks are used to minimize the time consumed in the simulation. The boundary conditions for mesh block 1 are as follows: The inlet and sides boundaries are defined as a wall boundary condition (wall boundary condition is usually used for bound fluid by solid regions. In the case of viscous flows, no-slip means that the tangential velocity is equal to the wall velocity and the normal velocity is zero), the outlet is defined as a symmetry boundary condition (symmetry boundary condition is usually used to reduce computational effort during CFD simulation. This condition allows the flow to be transferred from one mesh block to another. No inputs are required for this boundary condition except that its location should be defined accurately), the bottom boundary is defined as a uniform flow rate boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure. The boundary conditions for mesh block 2 are as follows: The inlet is defined as a symmetry boundary condition, the outlet is defined as a free flow boundary condition, the bottom and sides boundaries are defined as a wall boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure as shown in Fig. 2. The initial conditions required to be set for the fluid (i.e., water) inside of the domain include configuration, temperature, velocities, and pressure distribution. The configuration of water depends on the dimensions and shape of the dam reservoir. While the other conditions have been assigned as follows: temperature is normal water temperature (25 °c) and pressure distribution is hydrostatic with no initial velocity.

2.4. Numerical method

FLOW-3D uses the finite volume method (FVM) to solve the governing equation (Reynolds-averaged Navier-Stokes) over the computational domain. A finite-volume method is an Eulerian approach for representing and evaluating partial differential equations in algebraic equations form [29]. At discrete points on the mesh geometry, values are determined. Finite volume expresses a small volume surrounding each node point on a mesh. In this method, the divergence theorem is used to convert volume integrals with a divergence term to surface integrals. After that, these terms are evaluated as fluxes at each finite volume’s surfaces.

2.5. Turbulent models

Turbulence is the chaotic, unstable motion of fluids that occurs when there are insufficient stabilizing viscous forces. In FLOW-3D, there are six turbulence models available: the Prandtl mixing length model, the one-equation turbulent energy model, the two-equation (k – ε) model, the Renormalization-Group (RNG) model, the two-equation (k – ω) models, and a large eddy simulation (LES) model. For simulating flow motion, the RNG model is adopted to simulate the motion behavior better than the k – ε and k – ω.

models [30]. The RNG model consists of two main equations for the turbulent kinetic energy KT and its dissipation.εT(5)∂kT∂t+1VFuAx∂kT∂x+vAy∂kT∂y+wAz∂kT∂z=PT+GT+DiffKT-εT(6)∂εT∂t+1VFuAx∂εT∂x+vAy∂εT∂y+wAz∂εT∂z=C1.εTKTPT+c3.GT+Diffε-c2εT2kT

where KT is the turbulent kinetic energy, PT is the turbulent kinetic energy production, GT is the buoyancy turbulence energy, εT is the turbulent energy dissipation rate, DiffKT and Diffε are terms of diffusion, c1, c2 and c3 are dimensionless parameters, in which c1 and c3 have a constant value of 1.42 and 0.2, respectively, c2 is computed from the turbulent kinetic energy (KT) and turbulent production (PT) terms.

2.6. Sediment scour model

The sediment scour model available in FLOW-3D can calculate all the sediment transport processes including Entrainment transport, Bedload transport, Suspended transport, and Deposition. The erosion process starts once the water flows remove the grains from the packed bed and carry them into suspension. It happens when the applied shear stress by water flows exceeds critical shear stress. This process is represented by entrainment transport in the numerical model. After entrained, the grains carried by water flow are represented by suspended load transport. After that, some suspended grains resort to settling because of the combined effect of gravity, buoyancy, and friction. This process is described through a deposition. Finally, the grains sliding motions are represented by bedload transport in the model. For the entrainment process, the shear stress applied by the fluid motion on the packed bed surface is calculated using the standard wall function as shown in Eq.7.(7)ks,i=Cs,i∗d50

where ks,i is the Nikuradse roughness and Cs,i is a user-defined coefficient. The critical bed shear stress is defined by a dimensionless parameter called the critical shields number as expressed in Eq.8.(8)θcr,i=τcr,i‖g‖diρi-ρf

where θcr,i is the critical shields number, τcr,i is the critical bed shear stress, g is the absolute value of gravity acceleration, di is the diameter of the sediment grain, ρi is the density of the sediment species (i) and ρf is the density of the fluid. The value of the critical shields number is determined according to the Soulsby-Whitehouse equation.(9)θcr,i=0.31+1.2d∗,i+0.0551-exp-0.02d∗,i

where d∗,i is the dimensionless diameter of the sediment, given by Eq.10.(10)d∗,i=diρfρi-ρf‖g‖μf213

where μf is the fluid dynamic viscosity. For the sloping bed interface, the value of the critical shields number is modified according to Eq.11.(11)θ`cr,i=θcr,icosψsinβ+cos2βtan2φi-sin2ψsin2βtanφi

where θ`cr,i is the modified critical shields number, φi is the angle of repose for the sediment, β is the angle of bed slope and ψ is the angle between the flow and the upslope direction. The effects of the rolling, hopping, and sliding motions of grains along the packed bed surface are taken by the bedload transport process. The volumetric bedload transport rate (qb,i) per width of the bed is expressed in Eq.12.(12)qb,i=Φi‖g‖ρi-ρfρfdi312

where Φi is the dimensionless bedload transport rate is calculated by using Meyer Peter and Müller equation.(13)Φi=βMPM,iθi-θ`cr,i1.5cb,i

where βMPM,i is the Meyer Peter and Müller user-defined coefficient and cb,i is the volume fraction of species i in the bed material. The suspended load transport is calculated as shown in Eq.14.(14)∂Cs,i∂t+∇∙Cs,ius,i=∇∙∇DCs,i

where Cs,i is the suspended sediment mass concentration, D is the diffusivity, and us,i is the grain velocity of species i. Entrainment and deposition are two opposing processes that take place at the same time. The lifting and settling velocities for both entrainment and deposition processes are calculated according to Eq.15 and Eq.16, respectively.(15)ulifting,i=αid∗,i0.3θi-θ`cr,igdiρiρf-1(16)usettling,i=υfdi10.362+1.049d∗,i3-10.36

where αi is the entrainment coefficient of species i and υf is the kinematic viscosity of the fluid.

2.7. Grid type

Using simple rectangular orthogonal elements in planes and hexahedral in volumes in the (FLOW-3D) program makes the mesh generation process easier, decreases the required memory, and improves numerical accuracy. Two mesh blocks were used in a joined form with a size ratio of 2:1. The first mesh block is coarser, which contains the reservoir water, and the second mesh block is finer, which contains the dam. For achieving accuracy and efficiency in results, the mesh size is determined by using a grid convergence test. The optimum uniform cell size for the first mesh block is 0.012 m and for the second mesh block is 0.006 m.

2.8. Time step

The maximum time step size is determined by using a Courant number, which controls the distance that the flow will travel during the simulation time step. In this study, the Courant number was taken equal to 0.25 to prevent the flow from traveling through more than one cell in the time step. Based on the Courant number, a maximum time step value of 0.00075 s was determined.

2.9. Numerical model validation

The numerical model accuracy was achieved by comparing the numerical model results with previous experimental results. The experimental study of Schmocker and Hager [7] was based on 31 tests with changes in six parameters (d50, Ho, Bo, Lk, XD, and Qin). All experimental tests were conducted in a straight open glass-sided flume. The horizontal flume has a rectangular cross-section with a width of 0.4 m and a height of 0.7 m. The flume was provided with a flow straightener and an intake with a length of 0.66 m. All tested dams were inserted at various distances (XD) from the intake. Test No.1 from this experimental program was chosen to validate the numerical model. The different parameters used in test No.1 are as follows:

(1) uniform sediment with a mean diameter (d50 = 0.31 mm), (2) Ho = 0.2 m, (3) Bo = 0.2 m, (4) Lk = 0.1 m,

(5) XD = 1.0 m, (6) Qin = 6.0 lit/s, (7) Su and Sd = 2:1, (8) mass density (ρs = 2650 kg/m3(9) Homogenous and non-cohesive embankment dam. As shown in Fig. 2, the simulation is contained within a rectangular grid with dimensions: 3.56 m in the x-direction (where 0.66 m is used as inlet, 0.9 m as dam base width, and 1.0 m as outlet), in y-direction 0.2 m (dam length), and in the z-direction 0.3 m, which represents the dam height (0.2 m) with a free distance (0.1 m) above the dam. There are two main reasons that this experimental program is preferred for the validation process. The first reason is that this program deals with homogenous, non-cohesive soil, which is available in FLOW-3D. The second reason is that this program deals with small-scale models which saves time for numerical simulation. Finally, some important assumptions were considered during the validation process. The flow is assumed to be incompressible, viscous, turbulent, and three-dimensional.

By comparing dam profiles at different time instants for the experimental test with the current numerical model, it appears that the numerical model gives good agreement as shown in Fig. 3 and Fig. 4, with an average error percentage of 9% between the experimental results and the numerical model.

3. Analysis and discussions

The current model is used to study the effects of different parameters such as (initial breach shapes, dimensions, locations, upstream and downstream dam slopes) on the peak outflow discharge, QP, time of peak outflow, tP, and rate of erosion, E.

This study consists of a group of scenarios. The first scenario is changing the shapes of the initial breach according to Singh [1], the most predicted shapes are rectangular and V-notch as shown in Fig. 5. The second scenario is changing the initial breach dimensions (i.e., width and depth). While the third scenario is changing the location of the initial breach. Eventually, the last scenario is changing the upstream and downstream dam slopes.

All scenarios of this study were carried out under the same conditions such as inflow discharge value (Qin=1.0lit/s), dimensions of the tested dam, where dam height (Ho=0.20m), crest width.

(Lk=0.1m), dam length (Bo=0.20m), and homogenous & non-cohesive soil with a mean diameter (d50=0.31mm).

3.1. Dam breaching process evolution

The dam breaching process is a very complex process due to the quick changes in hydrodynamic conditions during dam failure. The dam breaching process starts once water flows reach the downstream face of the dam. During the initial stage of dam breaching, the erosion process is relatively quiet due to low velocities of flow. As water flows continuously, erosion rates increase, especially in two main zones: the crest and the downstream face. As soon as the dam crest is totally eroded, the water levels in the dam reservoir decrease rapidly, accompanied by excessive erosion in the dam body. The erosion process continues until the water levels in the dam reservoir equal the remaining height of the dam.

According to Zhou et al. [11], the breaching process consists of three main stages. The first stage starts with beginning overtopping flow, then ends when the erosion point directed upstream and reached the inflection point at the inflection time (ti). The second stage starts from the end of the stage1 until the occurrence of peak outflow discharge at the peak outflow time (tP). The third stage starts from the end of the stage2 until the value of outflow discharge becomes the same as the value of inflow discharge at the final time (tf). The outflow discharge from the dam breach increases rapidly during stage1 and stage2 because of the large dam storage capacity (i.e., the dam reservoir is totally full of water) and excessive erosion. While at stage3, the outflow values start to decrease slowly because most of the dam’s storage capacity was run out. The end of stage3 indicates that the dam storage capacity was totally run out, so the outflow equalized with the inflow discharge as shown in Fig. 6 and Fig. 7.

3.2. The effect of initial breach shape

To identify the effect of the initial breach shape on the evolution of the dam breaching process. Three tests were carried out with different cross-section areas for each shape. The initial breach is created at the center of the dam crest. Each test had an ID to make the process of arranging data easier. The rectangular shape had an ID (Rec5h & 5b), which means that its depth and width are equal to 5% of the dam height, and the V-notch shape had an ID (V-noch5h & 1:1) which means that its depth is equal to 5% of the dam height and its side slope is equal to 1:1. The comparison between rectangular and V-notch shapes is done by calculating the ratio between maximum dam height at different times (ZMax) to the initial dam height (Ho), rate of erosion, and hydrograph of outflow discharge for each test. The rectangular shape achieves maximum erosion rate and minimum inflection time, in addition to a rapid decrease in the dam reservoir levels. Therefore, the dam breaching is faster in the case of a rectangular shape than in a V-notch shape, which has the same cross-section area as shown in Fig. 8.

Also, by comparing the hydrograph for each test, the peak outflow discharge value in the case of a rectangular shape is higher than the V-notch shape by 5% and the time of peak outflow for the rectangular shape is shorter than the V-notch shape by 9% as shown in Fig. 9.

3.3. The effect of initial breach dimensions

The results of the comparison between the different initial breach shapes indicate that the worst initial breach shape is rectangular, so the second scenario from this study concentrated on studying the effect of a change in the initial rectangular breach dimensions. Groups of tests were carried out with different depths and widths for the rectangular initial breach. The first group had a depth of 5% from the dam height and with three different widths of 5,10, and 15% from the dam height, the second group had a depth of 10% with three different widths of 5,10, and 15%, the third group had a depth of 15% with three different widths of 5,10, and 15% and the final group had a width of 15% with three different heights of 5, 10, and 15% for a rectangular breach shape. The comparison was made as in the previous section to determine the worst case that leads to the quick dam failure as shown in Fig. 10.

The results show that the (Rec 5 h&15b) test achieves a maximum erosion rate for a shorter period of time and a minimum ratio for (Zmax / Ho) as shown in Fig. 10, which leads to accelerating the dam failure process. The dam breaching process is faster with the minimum initial breach depth and maximum initial breach width. In the case of a minimum initial breach depth, the retained head of water in the dam reservoir is high and the crest width at the bottom of the initial breach (L`K) is small, so the erosion point reaches the inflection point rapidly. While in the case of the maximum initial breach width, the erosion perimeter is large.

3.4. The effect of initial breach location

The results of the comparison between the different initial rectangular breach dimensions indicate that the worst initial breach dimension is (Rec 5 h&15b), so the third scenario from this study concentrated on studying the effect of a change in the initial breach location. Three locations were checked to determine the worst case for the dam failure process. The first location is at the center of the dam crest, which was named “Center”, the second location is at mid-distance between the dam center and dam edge, which was named “Mid”, and the third location is at the dam edge, which was named “Edge” as shown in Fig. 11. According to this scenario, the results indicate that the time of peak outflow discharge (tP) is the same in the three cases, but the maximum value of the peak outflow discharge occurs at the center location. The difference in the peak outflow values between the three cases is relatively small as shown in Fig. 12.

The rates of erosion were also studied for the three cases. The results show that the maximum erosion rate occurs at the center location as shown in Fig. 13. By making a comparison between the three cases for the dam storage volume. The results show that the center location had the minimum values for the dam storage volume, which means that a large amount of water has passed to the downstream area as shown in Fig. 14. According to these results, the center location leads to increased erosion rate and accelerated dam failure process compared with the two other cases. Because the erosion occurs on both sides, but in the case of edge location, the erosion occurs on one side.

3.5. The effect of upstream and downstream dam slopes

The results of the comparison between the different initial rectangular breach locations indicate that the worst initial breach location is the center location, so the fourth scenario from this study concentrated on studying the effect of a change in the upstream (Su) and downstream (Sd) dam slopes. Three slopes were checked individually for both upstream and downstream slopes to determine the worst case for the dam failure process. The first slope value is (2H:1V), the second slope value is (2.5H:1V), and the third slope value is (3H:1V). According to this scenario, the results show that the decreasing downstream slope angle leads to increasing time of peak outflow discharge (tP) and decreasing value of peak outflow discharge. The difference in the peak outflow values between the three cases for the downstream slope is 2%, as shown in Fig. 15, but changing the upstream slope has a negligible impact on the peak outflow discharge and its time as shown in Fig. 16.

The rates of erosion were also studied in the three cases for both upstream and downstream slopes. The results show that the maximum erosion rate increases by 6.0% with an increasing downstream slope angle by 4°, as shown in Fig. 17. The results also indicate that the erosion rates aren’t affected by increasing or decreasing the upstream slope angle, as shown in Fig. 18. According to these results, increasing the downstream slope angle leads to increased erosion rate and accelerated dam failure process compared with the upstream slope angle. Because of increasing shear stress applied by water flows in case of increasing downstream slope.

According to all previous scenarios, the dimensionless peak outflow discharge QPQin is presented for a fixed dam height (Ho) and inflow discharge (Qin). Fig. 19 illustrates the relationship between QP∗=QPQin and.

Lr=ho2/3∗bo2/3Ho. The deduced relationship achieves R2=0.96.(17)QP∗=2.2807exp-2.804∗Lr

4. Conclusions

A spatial dam breaching process was simulated by using FLOW-3D Software. The validation process was performed by making a comparison between the simulated results of dam profiles and the dam profiles obtained by Schmocker and Hager [7] in their experimental study. And also, the peak outflow value recorded an error percentage of 12% between the numerical model and the experimental study. This model was used to study the effect of initial breach shape, dimensions, location, and dam slopes on peak outflow discharge, time of peak outflow, and the erosion process. By using the parameters obtained from the validation process, the results of this study can be summarized in eight points as follows.1.

The rectangular initial breach shape leads to an accelerating dam failure process compared with the V-notch.2.

The value of peak outflow discharge in the case of a rectangular initial breach is higher than the V-notch shape by 5%.3.

The time of peak outflow discharge for a rectangular initial breach is shorter than the V-notch shape by 9%.4.

The minimum depth and maximum width for the initial breach achieve maximum erosion rates (increasing breach width, b0, or decreasing breach depth, h0, by 5% from the dam height leads to an increase in the maximum rate of erosion by 11% and 15%, respectively), so the dam failure is rapid.5.

The center location of the initial breach leads to an accelerating dam failure compared with the edge location.6.

The initial breach location has a negligible effect on the peak outflow discharge value and its time.7.

Increasing the downstream slope angle by 4° leads to an increase in both peak outflow discharge and maximum rate of erosion by 2.0% and 6.0%, respectively.8.

The upstream slope has a negligible effect on the dam breaching process.

References

[1]V. SinghDam breach modeling technologySpringer Science & Business Media (1996)Google Scholar[2]Wahl TL. Prediction of embankment dam breach parameters: a literature review and needs assessment. 1998.Google Scholar[3]Z. Alhasan, J. Jandora, J. ŘíhaStudy of dam-break due to overtopping of four small dams in the Czech RepublicActa Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63 (3) (2015), pp. 717-729 View PDFCrossRefView Record in ScopusGoogle Scholar[4]D. FreadBREACH, an erosion model for earthen dam failures: Hydrologic Research LaboratoryNOAA, National Weather Service (1988)Google Scholar[5]J. Říha, S. Kotaška, L. PetrulaDam Break Modeling in a Cascade of Small Earthen Dams: Case Study of the Čižina River in the Czech RepublicWater, 12 (8) (2020), p. 2309, 10.3390/w12082309 View PDFView Record in ScopusGoogle Scholar[6]E. Goodarzi, L. Teang Shui, M. ZiaeiDam overtopping risk using probabilistic concepts–Case study: The Meijaran DamIran Ain Shams Eng J, 4 (2) (2013), pp. 185-197ArticleDownload PDFView Record in ScopusGoogle Scholar[7]L. Schmocker, W.H. HagerPlane dike-breach due to overtopping: effects of sediment, dike height and dischargeJ Hydraul Res, 50 (6) (2012), pp. 576-586 View PDFCrossRefView Record in ScopusGoogle Scholar[8]J.S. Walder, R.M. Iverson, J.W. Godt, M. Logan, S.A. SolovitzControls on the breach geometry and flood hydrograph during overtopping of noncohesive earthen damsWater Resour Res, 51 (8) (2015), pp. 6701-6724View Record in ScopusGoogle Scholar[9]H. Wei, M. Yu, D. Wang, Y. LiOvertopping breaching of river levees constructed with cohesive sedimentsNat Hazards Earth Syst Sci, 16 (7) (2016), pp. 1541-1551 View PDFCrossRefView Record in ScopusGoogle Scholar[10]Y. Yang, S.-Y. Cao, K.-J. Yang, W.-P. LiYang K-j, Li W-p. Experimental study of breach process of landslide dams by overtopping and its initiation mechanismsJ Hydrodynamics, 27 (6) (2015), pp. 872-883ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[11]G.G.D. Zhou, M. Zhou, M.S. Shrestha, D. Song, C.E. Choi, K.F.E. Cui, et al.Experimental investigation on the longitudinal evolution of landslide dam breaching and outburst floodsGeomorphology, 334 (2019), pp. 29-43ArticleDownload PDFView Record in ScopusGoogle Scholar[12]J. Zhang, Z.-x. Guo, S.-y. CaoYang F-g. Experimental study on scour and erosion of blocked damWater Sci Eng, 5 (2012), pp. 219-229ArticleDownload PDFView Record in ScopusGoogle Scholar[13]K. Höeg, A. Løvoll, K. VaskinnStability and breaching of embankment dams: Field tests on 6 m high damsInt J Hydropower Dams, 11 (2004), pp. 88-92View Record in ScopusGoogle Scholar[14]H. Hakimzadeh, V. Nourani, A.B. AminiGenetic programming simulation of dam breach hydrograph and peak outflow dischargeJ Hydrol Eng, 19 (4) (2014), pp. 757-768View Record in ScopusGoogle Scholar[15]A.R. Refaiy, N.M. AboulAtta, N.Y. Saad, D.A. El-MollaModeling the effect of downstream drain geometry on seepage through earth damsAin Shams Eng J, 12 (3) (2021), pp. 2511-2531ArticleDownload PDFView Record in ScopusGoogle Scholar[16]Y. Zhu, P.J. Visser, J.K. Vrijling, G. WangExperimental investigation on breaching of embankmentsScience China Technological Sci, 54 (1) (2011), pp. 148-155 View PDFCrossRefView Record in ScopusGoogle Scholar[17]M.-H. Yu, H.-Y. Wei, Y.-J. Liang, Y. ZhaoInvestigation of non-cohesive levee breach by overtopping flowJ Hydrodyn, 25 (4) (2013), pp. 572-579ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[18]S. Wu, M. Yu, H. Wei, Y. Liang, J. ZengNon-symmetrical levee breaching processes in a channel bend due to overtoppingInt J Sedim Res, 33 (2) (2018), pp. 208-215ArticleDownload PDFView Record in ScopusGoogle Scholar[19]O. Saberi, G. ZenzNumerical investigation on 1D and 2D embankment dams failure due to overtopping flowInt J Hydraulic Engineering, 5 (2016), pp. 9-18View Record in ScopusGoogle Scholar[20]M. Guan, N.G. Wright, P.A. Sleigh2D Process-Based Morphodynamic Model for Flooding by Noncohesive Dyke BreachJ Hydraul Eng, 140 (7) (2014), p. 04014022, 10.1061/(ASCE)HY.1943-7900.0000861 View PDFView Record in ScopusGoogle Scholar[21]W. Wu, R. Marsooli, Z. HeDepth-Averaged Two-Dimensional Model of Unsteady Flow and Sediment Transport due to Noncohesive Embankment Break/BreachingJ Hydraul Eng, 138 (6) (2012), pp. 503-516View Record in ScopusGoogle Scholar[22]Z. Wang, D.S. BowlesThree-dimensional non-cohesive earthen dam breach model. Part 1: Theory and methodologyAdv Water Resour, 29 (10) (2006), pp. 1528-1545ArticleDownload PDFView Record in ScopusGoogle Scholar[23]Říha J, Duchan D, Zachoval Z, Erpicum S, Archambeau P, Pirotton M, et al. Performance of a shallow-water model for simulating flow over trapezoidal broad-crested weirs. J Hydrology Hydromechanics. 2019;67:322-8.Google Scholar[24]C.B. VreugdenhilNumerical methods for shallow-water flowSpringer Science & Business Media (1994)Google Scholar[25]L.A. Larocque, J. Imran, M.H. Chaudhry3D numerical simulation of partial breach dam-break flow using the LES and k–∊ turbulence modelsJ Hydraul Res, 51 (2) (2013), pp. 145-157 View PDFCrossRefView Record in ScopusGoogle Scholar[26]C. Yang, B. Lin, C. Jiang, Y. LiuPredicting near-field dam-break flow and impact force using a 3D modelJ Hydraul Res, 48 (6) (2010), pp. 784-792 View PDFCrossRefView Record in ScopusGoogle Scholar[27]FLOW-3D. Version 11.1.1 Flow Science, Inc., Santa Fe, NM. https://wwwflow3dcom.Google Scholar[28]C.W. Hirt, B.D. NicholsVolume of fluid (VOF) method for the dynamics of free boundariesJ Comput Phys, 39 (1) (1981), pp. 201-225ArticleDownload PDFGoogle Scholar[29]S.V. PatankarNumerical heat transfer and fluid flow, Hemisphere PublCorp, New York, 58 (1980), p. 288View Record in ScopusGoogle Scholar[30]M. Alemi, R. MaiaNumerical simulation of the flow and local scour process around single and complex bridge piersInt J Civil Eng, 16 (5) (2018), pp. 475-487 View PDFCrossRefView Record in ScopusGoogle Scholar

Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

다공성 미디어 및 나노유체에 의해 강화된 수집기로 태양광 CCHP 시스템의 최적화

Optimization of Solar CCHP Systems with Collector Enhanced by Porous Media and Nanofluid


Navid Tonekaboni,1Mahdi Feizbahr,2 Nima Tonekaboni,1Guang-Jun Jiang,3,4 and Hong-Xia Chen3,4

Abstract

태양열 집열기의 낮은 효율은 CCHP(Solar Combined Cooling, Heating, and Power) 사이클의 문제점 중 하나로 언급될 수 있습니다. 태양계를 개선하기 위해 나노유체와 다공성 매체가 태양열 집열기에 사용됩니다.

다공성 매질과 나노입자를 사용하는 장점 중 하나는 동일한 조건에서 더 많은 에너지를 흡수할 수 있다는 것입니다. 이 연구에서는 평균 일사량이 1b인 따뜻하고 건조한 지역의 600 m2 건물의 전기, 냉방 및 난방을 생성하기 위해 다공성 매질과 나노유체를 사용하여 태양열 냉난방 복합 발전(SCCHP) 시스템을 최적화했습니다.

본 논문에서는 침전물이 형성되지 않는 lb = 820 w/m2(이란) 정도까지 다공성 물질에서 나노유체의 최적량을 계산하였다. 이 연구에서 태양열 집열기는 구리 다공성 매체(95% 다공성)와 CuO 및 Al2O3 나노 유체로 향상되었습니다.

나노유체의 0.1%-0.6%가 작동 유체로 물에 추가되었습니다. 나노유체의 0.5%가 태양열 집열기 및 SCCHP 시스템에서 가장 높은 에너지 및 엑서지 효율 향상으로 이어지는 것으로 밝혀졌습니다.

본 연구에서 포물선형 집열기(PTC)의 최대 에너지 및 엑서지 효율은 각각 74.19% 및 32.6%입니다. 그림 1은 태양 CCHP의 주기를 정확하게 설명하기 위한 그래픽 초록으로 언급될 수 있습니다.

The low efficiency of solar collectors can be mentioned as one of the problems in solar combined cooling, heating, and power (CCHP) cycles. For improving solar systems, nanofluid and porous media are used in solar collectors. One of the advantages of using porous media and nanoparticles is to absorb more energy under the same conditions. In this research, a solar combined cooling, heating, and power (SCCHP) system has been optimized by porous media and nanofluid for generating electricity, cooling, and heating of a 600 m2 building in a warm and dry region with average solar radiation of Ib = 820 w/m2 in Iran. In this paper, the optimal amount of nanofluid in porous materials has been calculated to the extent that no sediment is formed. In this study, solar collectors were enhanced with copper porous media (95% porosity) and CuO and Al2O3 nanofluids. 0.1%–0.6% of the nanofluids were added to water as working fluids; it is found that 0.5% of the nanofluids lead to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Maximum energy and exergy efficiency of parabolic thermal collector (PTC) riches in this study are 74.19% and 32.6%, respectively. Figure 1 can be mentioned as a graphical abstract for accurately describing the cycle of solar CCHP.

1. Introduction

Due to the increase in energy consumption, the use of clean energy is one of the important goals of human societies. In the last four decades, the use of cogeneration cycles has increased significantly due to high efficiency. Among clean energy, the use of solar energy has become more popular due to its greater availability [1]. Low efficiency of energy production, transmission, and distribution system makes a new system to generate simultaneously electricity, heating, and cooling as an essential solution to be widely used. The low efficiency of the electricity generation, transmission, and distribution system makes the CCHP system a basic solution to eliminate waste of energy. CCHP system consists of a prime mover (PM), a power generator, a heat recovery system (produce extra heating/cooling/power), and thermal energy storage (TES) [2]. Solar combined cooling, heating, and power (SCCHP) has been started three decades ago. SCCHP is a system that receives its propulsive force from solar energy; in this cycle, solar collectors play the role of propulsive for generating power in this system [3].

Increasing the rate of energy consumption in the whole world because of the low efficiency of energy production, transmission, and distribution system causes a new cogeneration system to generate electricity, heating, and cooling energy as an essential solution to be widely used. Building energy utilization fundamentally includes power required for lighting, home electrical appliances, warming and cooling of building inside, and boiling water. Domestic usage contributes to an average of 35% of the world’s total energy consumption [4].

Due to the availability of solar energy in all areas, solar collectors can be used to obtain the propulsive power required for the CCHP cycle. Solar energy is the main source of energy in renewable applications. For selecting a suitable area to use solar collectors, annual sunshine hours, the number of sunny days, minus temperature and frosty days, and the windy status of the region are essentially considered [5]. Iran, with an average of more than 300 sunny days, is one of the suitable countries to use solar energy. Due to the fact that most of the solar radiation is in the southern regions of Iran, also the concentration of cities is low in these areas, and transmission lines are far apart, one of the best options is to use CCHP cycles based on solar collectors [6]. One of the major problems of solar collectors is their low efficiency [7]. Low efficiency increases the area of collectors, which increases the initial cost of solar systems and of course increases the initial payback period. To increase the efficiency of solar collectors and improve their performance, porous materials and nanofluids are used to increase their workability.

There are two ways to increase the efficiency of solar collectors and mechanical and fluid improvement. In the first method, using porous materials or helical filaments inside the collector pipes causes turbulence of the flow and increases heat transfer. In the second method, using nanofluids or salt and other materials increases the heat transfer of water. The use of porous materials has grown up immensely over the past twenty years. Porous materials, especially copper porous foam, are widely used in solar collectors. Due to the high contact surface area, porous media are appropriate candidates for solar collectors [8]. A number of researchers investigated Solar System performance in accordance with energy and exergy analyses. Zhai et al. [9] reviewed the performance of a small solar-powered system in which the energy efficiency was 44.7% and the electrical efficiency was 16.9%.

Abbasi et al. [10] proposed an innovative multiobjective optimization to optimize the design of a cogeneration system. Results showed the CCHP system based on an internal diesel combustion engine was the applicable alternative at all regions with different climates. The diesel engine can supply the electrical requirement of 31.0% and heating demand of 3.8% for building.

Jiang et al. [11] combined the experiment and simulation together to analyze the performance of a cogeneration system. Moreover, some research focused on CCHP systems using solar energy. It integrated sustainable and renewable technologies in the CCHP, like PV, Stirling engine, and parabolic trough collector (PTC) [21215].

Wang et al. [16] optimized a cogeneration solar cooling system with a Rankine cycle and ejector to reach the maximum total system efficiency of 55.9%. Jing et al. analyzed a big-scale building with the SCCHP system and auxiliary heaters to produced electrical, cooling, and heating power. The maximum energy efficiency reported in their work is 46.6% [17]. Various optimization methods have been used to improve the cogeneration system, minimum system size, and performance, such as genetic algorithm [1819].

Hirasawa et al. [20] investigated the effect of using porous media to reduce thermal waste in solar systems. They used the high-porosity metal foam on top of the flat plate solar collector and observed that thermal waste decreased by 7% due to natural heat transfer. Many researchers study the efficiency improvement of the solar collector by changing the collector’s shapes or working fluids. However, the most effective method is the use of nanofluids in the solar collector as working fluid [21]. In the experimental study done by Jouybari et al. [22], the efficiency enhancement up to 8.1% was achieved by adding nanofluid in a flat plate collector. In this research, by adding porous materials to the solar collector, collector efficiency increased up to 92% in a low flow regime. Subramani et al. [23] analyzed the thermal performance of the parabolic solar collector with Al2O3 nanofluid. They conducted their experiments with Reynolds number range 2401 to 7202 and mass flow rate 0.0083 to 0.05 kg/s. The maximum efficiency improvement in this experiment was 56% at 0.05 kg/s mass flow rate.

Shojaeizadeh et al. [24] investigated the analysis of the second law of thermodynamic on the flat plate solar collector using Al2O3/water nanofluid. Their research showed that energy efficiency rose up to 1.9% and the exergy efficiency increased by a maximum of 0.72% compared to pure water. Tiwari et al. [25] researched on the thermal performance of solar flat plate collectors for working fluid water with different nanofluids. The result showed that using 1.5% (optimum) particle volume fraction of Al2O3 nanofluid as an absorbing medium causes the thermal efficiency to enhance up to 31.64%.

The effect of porous media and nanofluids on solar collectors has already been investigated in the literature but the SCCHP system with a collector embedded by both porous media and nanofluid for enhancing the ratio of nanoparticle in nanofluid for preventing sedimentation was not discussed. In this research, the amount of energy and exergy of the solar CCHP cycles with parabolic solar collectors in both base and improved modes with a porous material (copper foam with 95% porosity) and nanofluid with different ratios of nanoparticles was calculated. In the first step, it is planned to design a CCHP system based on the required load, and, in the next step, it will analyze the energy and exergy of the system in a basic and optimize mode. In the optimize mode, enhanced solar collectors with porous material and nanofluid in different ratios (0.1%–0.7%) were used to optimize the ratio of nanofluids to prevent sedimentation.

2. Cycle Description

CCHP is one of the methods to enhance energy efficiency and reduce energy loss and costs. The SCCHP system used a solar collector as a prime mover of the cogeneration system and assisted the boiler to generate vapor for the turbine. Hot water flows from the expander to the absorption chiller in summer or to the radiator or fan coil in winter. Finally, before the hot water wants to flow back to the storage tank, it flows inside a heat exchanger for generating domestic hot water [26].

For designing of solar cogeneration system and its analysis, it is necessary to calculate the electrical, heating (heating load is the load required for the production of warm water and space heating), and cooling load required for the case study considered in a residential building with an area of 600 m2 in the warm region of Iran (Zahedan). In Table 1, the average of the required loads is shown for the different months of a year (average of electrical, heating, and cooling load calculated with CARRIER software).Table 1 The average amount of electric charges, heating load, and cooling load used in the different months of the year in the city of Zahedan for a residential building with 600 m2.

According to Table 1, the maximum magnitude of heating, cooling, and electrical loads is used to calculate the cogeneration system. The maximum electric load is 96 kW, the maximum amount of heating load is 62 kW, and the maximum cooling load is 118 kW. Since the calculated loads are average, all loads increased up to 10% for the confidence coefficient. With the obtained values, the solar collector area and other cogeneration system components are calculated. The cogeneration cycle is capable of producing 105 kW electric power, 140 kW cooling capacity, and 100 kW heating power.

2.1. System Analysis Equations

An analysis is done by considering the following assumptions:(1)The system operates under steady-state conditions(2)The system is designed for the warm region of Iran (Zahedan) with average solar radiation Ib = 820 w/m2(3)The pressure drops in heat exchangers, separators, storage tanks, and pipes are ignored(4)The pressure drop is negligible in all processes and no expectable chemical reactions occurred in the processes(5)Potential, kinetic, and chemical exergy are not considered due to their insignificance(6)Pumps have been discontinued due to insignificance throughout the process(7)All components are assumed adiabatic

Schematic shape of the cogeneration cycle is shown in Figure 1 and all data are given in Table 2.

Figure 1 Schematic shape of the cogeneration cycle.Table 2 Temperature and humidity of different points of system.

Based on the first law of thermodynamic, energy analysis is based on the following steps.

First of all, the estimated solar radiation energy on collector has been calculated:where α is the heat transfer enhancement coefficient based on porous materials added to the collector’s pipes. The coefficient α is increased by the porosity percentage, the type of porous material (in this case, copper with a porosity percentage of 95), and the flow of fluid to the collector equation.

Collector efficiency is going to be calculated by the following equation [9]:

Total energy received by the collector is given by [9]

Also, the auxiliary boiler heat load is [2]

Energy consumed from vapor to expander is calculated by [2]

The power output form by the screw expander [9]:

The efficiency of the expander is 80% in this case [11].

In this step, cooling and heating loads were calculated and then, the required heating load to reach sanitary hot water will be calculated as follows:

First step: calculating the cooling load with the following equation [9]:

Second step: calculating heating loads [9]:

Then, calculating the required loud for sanitary hot water will be [9]

According to the above-mentioned equations, efficiency is [9]

In the third step, calculated exergy analysis as follows.

First, the received exergy collector from the sun is calculated [9]:

In the previous equation, f is the constant of air dilution.

The received exergy from the collector is [9]

In the case of using natural gas in an auxiliary heater, the gas exergy is calculated from the following equation [12]:

Delivering exergy from vapor to expander is calculated with the following equation [9]:

In the fourth step, the exergy in cooling and heating is calculated by the following equation:

Cooling exergy in summer is calculated [9]:

Heating exergy in winter is calculated [9]:

In the last step based on thermodynamic second law, exergy efficiency has been calculated from the following equation and the above-mentioned calculated loads [9]:

3. Porous Media

The porous medium that filled the test section is copper foam with a porosity of 95%. The foams are determined in Figure 2 and also detailed thermophysical parameters and dimensions are shown in Table 3.

Figure 2 Copper foam with a porosity of 95%.Table 3 Thermophysical parameters and dimensions of copper foam.

In solar collectors, copper porous materials are suitable for use at low temperatures and have an easier and faster manufacturing process than ceramic porous materials. Due to the high coefficient conductivity of copper, the use of copper metallic foam to increase heat transfer is certainly more efficient in solar collectors.

Porous media and nanofluid in solar collector’s pipes were simulated in FLOW-3D software using the finite-difference method [27]. Nanoparticles Al2O3 and CUO are mostly used in solar collector enhancement. In this research, different concentrations of nanofluid are added to the parabolic solar collectors with porous materials (copper foam with porosity of 95%) to achieve maximum heat transfer in the porous materials before sedimentation. After analyzing PTC pipes with the nanofluid flow in FLOW-3D software, for energy and exergy efficiency analysis, Carrier software results were used as EES software input. Simulation PTC with porous media inside collector pipe and nanofluids sedimentation is shown in Figure 3.

Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

3.1. Nano Fluid

In this research, copper and silver nanofluids (Al2O3, CuO) have been added with percentages of 0.1%–0.7% as the working fluids. The nanoparticle properties are given in Table 4. Also, system constant parameters are presented in Table 4, which are available as default input in the EES software.Table 4 Properties of the nanoparticles [9].

System constant parameters for input in the software are shown in Table 5.Table 5 System constant parameters.

The thermal properties of the nanofluid can be obtained from equations (18)–(21). The basic fluid properties are indicated by the index (bf) and the properties of the nanoparticle silver with the index (np).

The density of the mixture is shown in the following equation [28]:where ρ is density and ϕ is the nanoparticles volume fraction.

The specific heat capacity is calculated from the following equation [29]:

The thermal conductivity of the nanofluid is calculated from the following equation [29]:

The parameter β is the ratio of the nanolayer thickness to the original particle radius and, usually, this parameter is taken equal to 0.1 for the calculated thermal conductivity of the nanofluids.

The mixture viscosity is calculated as follows [30]:

In all equations, instead of water properties, working fluids with nanofluid are used. All of the above equations and parameters are entered in the EES software for calculating the energy and exergy of solar collectors and the SCCHP cycle. All calculation repeats for both nanofluids with different concentrations of nanofluid in the solar collector’s pipe.

4. Results and Discussion

In the present study, relations were written according to Wang et al. [16] and the system analysis was performed to ensure the correctness of the code. The energy and exergy charts are plotted based on the main values of the paper and are shown in Figures 4 and 5. The error rate in this simulation is 1.07%.

Figure 4 Verification charts of energy analysis results.

Figure 5 Verification charts of exergy analysis results.

We may also investigate the application of machine learning paradigms [3141] and various hybrid, advanced optimization approaches that are enhanced in terms of exploration and intensification [4255], and intelligent model studies [5661] as well, for example, methods such as particle swarm optimizer (PSO) [6062], differential search (DS) [63], ant colony optimizer (ACO) [616465], Harris hawks optimizer (HHO) [66], grey wolf optimizer (GWO) [5367], differential evolution (DE) [6869], and other fusion and boosted systems [4146485054557071].

At the first step, the collector is modified with porous copper foam material. 14 cases have been considered for the analysis of the SCCHP system (Table 6). It should be noted that the adding of porous media causes an additional pressure drop inside the collector [922263072]. All fourteen cases use copper foam with a porosity of 95 percent. To simulate the effect of porous materials and nanofluids, the first solar PTC pipes have been simulated in the FLOW-3D software and then porous media (copper foam with porosity of 95%) and fluid flow with nanoparticles (AL2O3 and CUO) are generated in the software. After analyzing PTC pipes in FLOW-3D software, for analyzing energy and exergy efficiency, software outputs were used as EES software input for optimization ratio of sedimentation and calculating energy and exergy analyses.Table 6 Collectors with different percentages of nanofluids and porous media.

In this research, an enhanced solar collector with both porous media and Nanofluid is investigated. In the present study, 0.1–0.5% CuO and Al2O3 concentration were added to the collector fully filled by porous media to achieve maximum energy and exergy efficiencies of solar CCHP systems. All steps of the investigation are shown in Table 6.

Energy and exergy analyses of parabolic solar collectors and SCCHP systems are shown in Figures 6 and 7.

Figure 6 Energy and exergy efficiencies of the PTC with porous media and nanofluid.

Figure 7 Energy and exergy efficiency of the SCCHP.

Results show that the highest energy and exergy efficiencies are 74.19% and 32.6%, respectively, that is achieved in Step 12 (parabolic collectors with filled porous media and 0.5% Al2O3). In the second step, the maximum energy efficiency of SCCHP systems with fourteen steps of simulation are shown in Figure 7.

In the second step, where 0.1, −0.6% of the nanofluids were added, it is found that 0.5% leads to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Using concentrations more than 0.5% leads to sediment in the solar collector’s pipe and a decrease of porosity in the pipe [73]. According to Figure 7, maximum energy and exergy efficiencies of SCCHP are achieved in Step 12. In this step energy efficiency is 54.49% and exergy efficiency is 18.29%. In steps 13 and 14, with increasing concentration of CUO and Al2O3 nanofluid solution in porous materials, decreasing of energy and exergy efficiency of PTC and SCCHP system at the same time happened. This decrease in efficiency is due to the formation of sediment in the porous material. Calculations and simulations have shown that porous materials more than 0.5% nanofluids inside the collector pipe cause sediment and disturb the porosity of porous materials and pressure drop and reduce the coefficient of performance of the cogeneration system. Most experience showed that CUO and AL2O3 nanofluids with less than 0.6% percent solution are used in the investigation on the solar collectors at low temperatures and discharges [74]. One of the important points of this research is that the best ratio of nanofluids in the solar collector with a low temperature is 0.5% (AL2O3 and CUO); with this replacement, the cost of solar collectors and SCCHP cycle is reduced.

5. Conclusion and Future Directions

In the present study, ways for increasing the efficiency of solar collectors in order to enhance the efficiency of the SCCHP cycle are examined. The research is aimed at adding both porous materials and nanofluids for estimating the best ratio of nanofluid for enhanced solar collector and protecting sedimentation in porous media. By adding porous materials (copper foam with porosity of 95%) and 0.5% nanofluids together, high efficiency in solar parabolic collectors can be achieved. The novelty in this research is the addition of both nanofluids and porous materials and calculating the best ratio for preventing sedimentation and pressure drop in solar collector’s pipe. In this study, it was observed that, by adding 0.5% of AL2O3 nanofluid in working fluids, the energy efficiency of PTC rises to 74.19% and exergy efficiency is grown up to 32.6%. In SCCHP cycle, energy efficiency is 54.49% and exergy efficiency is 18.29%.

In this research, parabolic solar collectors fully filled by porous media (copper foam with a porosity of 95) are investigated. In the next step, parabolic solar collectors in the SCCHP cycle were simultaneously filled by porous media and different percentages of Al2O3 and CuO nanofluid. At this step, values of 0.1% to 0.6% of each nanofluid were added to the working fluid, and the efficiency of the energy and exergy of the collectors and the SCCHP cycle were determined. In this case, nanofluid and the porous media were used together in the solar collector and maximum efficiency achieved. 0.5% of both nanofluids were used to achieve the biggest efficiency enhancement.

In the present study, as expected, the highest efficiency is for the parabolic solar collector fully filled by porous material (copper foam with a porosity of 95%) and 0.5% Al2O3. Results of the present study are as follows:(1)The average enhancement of collectors’ efficiency using porous media and nanofluids is 28%.(2)Solutions with 0.1 to 0.5% of nanofluids (CuO and Al2O3) are used to prevent collectors from sediment occurrence in porous media.(3)Collector of solar cogeneration cycles that is enhanced by both porous media and nanofluid has higher efficiency, and the stability of output temperature is more as well.(4)By using 0.6% of the nanofluids in the enhanced parabolic solar collectors with copper porous materials, sedimentation occurs and makes a high-pressure drop in the solar collector’s pipe which causes decrease in energy efficiency.(5)Average enhancement of SCCHP cycle efficiency is enhanced by both porous media and nanofluid 13%.

Nomenclature

:Solar radiation
a:Heat transfer augmentation coefficient
A:Solar collector area
Bf:Basic fluid
:Specific heat capacity of the nanofluid
F:Constant of air dilution
:Thermal conductivity of the nanofluid
:Thermal conductivity of the basic fluid
:Viscosity of the nanofluid
:Viscosity of the basic fluid
:Collector efficiency
:Collector energy receives
:Auxiliary boiler heat
:Expander energy
:Gas energy
:Screw expander work
:Cooling load, in kilowatts
:Heating load, in kilowatts
:Solar radiation energy on collector, in Joule
:Sanitary hot water load
Np:Nanoparticle
:Energy efficiency
:Heat exchanger efficiency
:Sun exergy
:Collector exergy
:Natural gas exergy
:Expander exergy
:Cooling exergy
:Heating exergy
:Exergy efficiency
:Steam mass flow rate
:Hot water mass flow rate
:Specific heat capacity of water
:Power output form by the screw expander
Tam:Average ambient temperature
:Density of the mixture.

Greek symbols

ρ:Density
ϕ:Nanoparticles volume fraction
β:Ratio of the nanolayer thickness.

Abbreviations

CCHP:Combined cooling, heating, and power
EES:Engineering equation solver.

Data Availability

For this study, data were generated by CARRIER software for the average electrical, heating, and cooling load of a residential building with 600 m2 in the city of Zahedan, Iran.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was partially supported by the National Natural Science Foundation of China under Contract no. 71761030 and Natural Science Foundation of Inner Mongolia under Contract no. 2019LH07003.

References

  1. A. Fudholi and K. Sopian, “Review on solar collector for agricultural produce,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 414, 2018.View at: Publisher Site | Google Scholar
  2. G. Yang and X. Zhai, “Optimization and performance analysis of solar hybrid CCHP systems under different operation strategies,” Applied Thermal Engineering, vol. 133, pp. 327–340, 2018.View at: Publisher Site | Google Scholar
  3. J. Wang, Z. Han, and Z. Guan, “Hybrid solar-assisted combined cooling, heating, and power systems: a review,” Renewable and Sustainable Energy Reviews, vol. 133, p. 110256, 2020.View at: Publisher Site | Google Scholar
  4. Y. Tian and C. Y. Zhao, “A review of solar collectors and thermal energy storage in solar thermal applications,” Applied Energy, vol. 104, pp. 538–553, 2013.View at: Publisher Site | Google Scholar
  5. J. M. Hassan, Q. J. Abdul-Ghafour, and M. F. Mohammed, “CFD simulation of enhancement techniques in flat plate solar water collectors,” Al-Nahrain Journal for Engineering Sciences, vol. 20, no. 3, pp. 751–761, 2017.View at: Google Scholar
  6. M. Jahangiri, O. Nematollahi, A. Haghani, H. A. Raiesi, and A. Alidadi Shamsabadi, “An optimization of energy cost of clean hybrid solar-wind power plants in Iran,” International Journal of Green Energy, vol. 16, no. 15, pp. 1422–1435, 2019.View at: Publisher Site | Google Scholar
  7. I. H. Yılmaz and A. Mwesigye, “Modeling, simulation and performance analysis of parabolic trough solar collectors: a comprehensive review,” Applied Energy, vol. 225, pp. 135–174, 2018.View at: Google Scholar
  8. F. Wang, J. Tan, and Z. Wang, “Heat transfer analysis of porous media receiver with different transport and thermophysical models using mixture as feeding gas,” Energy Conversion and Management, vol. 83, pp. 159–166, 2014.View at: Publisher Site | Google Scholar
  9. H. Zhai, Y. J. Dai, J. Y. Wu, and R. Z. Wang, “Energy and exergy analyses on a novel hybrid solar heating, cooling and power generation system for remote areas,” Applied Energy, vol. 86, no. 9, pp. 1395–1404, 2009.View at: Publisher Site | Google Scholar
  10. M. H. Abbasi, H. Sayyaadi, and M. Tahmasbzadebaie, “A methodology to obtain the foremost type and optimal size of the prime mover of a CCHP system for a large-scale residential application,” Applied Thermal Engineering, vol. 135, pp. 389–405, 2018.View at: Google Scholar
  11. R. Jiang, F. G. F. Qin, X. Yang, S. Huang, and B. Chen, “Performance analysis of a liquid absorption dehumidifier driven by jacket-cooling water of a diesel engine in a CCHP system,” Energy and Buildings, vol. 163, pp. 70–78, 2018.View at: Publisher Site | Google Scholar
  12. F. A. Boyaghchi and M. Chavoshi, “Monthly assessments of exergetic, economic and environmental criteria and optimization of a solar micro-CCHP based on DORC,” Solar Energy, vol. 166, pp. 351–370, 2018.View at: Publisher Site | Google Scholar
  13. F. A. Boyaghchi and M. Chavoshi, “Multi-criteria optimization of a micro solar-geothermal CCHP system applying water/CuO nanofluid based on exergy, exergoeconomic and exergoenvironmental concepts,” Applied Thermal Engineering, vol. 112, pp. 660–675, 2017.View at: Publisher Site | Google Scholar
  14. B. Su, W. Han, Y. Chen, Z. Wang, W. Qu, and H. Jin, “Performance optimization of a solar assisted CCHP based on biogas reforming,” Energy Conversion and Management, vol. 171, pp. 604–617, 2018.View at: Publisher Site | Google Scholar
  15. F. A. Al-Sulaiman, F. Hamdullahpur, and I. Dincer, “Performance assessment of a novel system using parabolic trough solar collectors for combined cooling, heating, and power production,” Renewable Energy, vol. 48, pp. 161–172, 2012.View at: Publisher Site | Google Scholar
  16. J. Wang, Y. Dai, L. Gao, and S. Ma, “A new combined cooling, heating and power system driven by solar energy,” Renewable Energy, vol. 34, no. 12, pp. 2780–2788, 2009.View at: Publisher Site | Google Scholar
  17. Y.-Y. Jing, H. Bai, J.-J. Wang, and L. Liu, “Life cycle assessment of a solar combined cooling heating and power system in different operation strategies,” Applied Energy, vol. 92, pp. 843–853, 2012.View at: Publisher Site | Google Scholar
  18. J.-J. Wang, Y.-Y. Jing, and C.-F. Zhang, “Optimization of capacity and operation for CCHP system by genetic algorithm,” Applied Energy, vol. 87, no. 4, pp. 1325–1335, 2010.View at: Publisher Site | Google Scholar
  19. L. Ali, “LDA–GA–SVM: improved hepatocellular carcinoma prediction through dimensionality reduction and genetically optimized support vector machine,” Neural Computing and Applications, vol. 87, pp. 1–10, 2020.View at: Google Scholar
  20. S. Hirasawa, R. Tsubota, T. Kawanami, and K. Shirai, “Reduction of heat loss from solar thermal collector by diminishing natural convection with high-porosity porous medium,” Solar Energy, vol. 97, pp. 305–313, 2013.View at: Publisher Site | Google Scholar
  21. E. Bellos, C. Tzivanidis, and Z. Said, “A systematic parametric thermal analysis of nanofluid-based parabolic trough solar collectors,” Sustainable Energy Technologies and Assessments, vol. 39, p. 100714, 2020.View at: Publisher Site | Google Scholar
  22. H. J. Jouybari, S. Saedodin, A. Zamzamian, M. E. Nimvari, and S. Wongwises, “Effects of porous material and nanoparticles on the thermal performance of a flat plate solar collector: an experimental study,” Renewable Energy, vol. 114, pp. 1407–1418, 2017.View at: Publisher Site | Google Scholar
  23. J. Subramani, P. K. Nagarajan, S. Wongwises, S. A. El-Agouz, and R. Sathyamurthy, “Experimental study on the thermal performance and heat transfer characteristics of solar parabolic trough collector using Al2O3 nanofluids,” Environmental Progress & Sustainable Energy, vol. 37, no. 3, pp. 1149–1159, 2018.View at: Publisher Site | Google Scholar
  24. E. Shojaeizadeh, F. Veysi, and A. Kamandi, “Exergy efficiency investigation and optimization of an Al2O3-water nanofluid based Flat-plate solar collector,” Energy and Buildings, vol. 101, pp. 12–23, 2015.View at: Publisher Site | Google Scholar
  25. A. K. Tiwari, P. Ghosh, and J. Sarkar, “Solar water heating using nanofluids–a comprehensive overview and environmental impact analysis,” International Journal of Emerging Technology and Advanced Engineering, vol. 3, no. 3, pp. 221–224, 2013.View at: Google Scholar
  26. D. R. Rajendran, E. Ganapathy Sundaram, P. Jawahar, V. Sivakumar, O. Mahian, and E. Bellos, “Review on influencing parameters in the performance of concentrated solar power collector based on materials, heat transfer fluids and design,” Journal of Thermal Analysis and Calorimetry, vol. 140, no. 1, pp. 33–51, 2020.View at: Publisher Site | Google Scholar
  27. M. Feizbahr, C. Kok Keong, F. Rostami, and M. Shahrokhi, “Wave energy dissipation using perforated and non perforated piles,” International Journal of Engineering, vol. 31, no. 2, pp. 212–219, 2018.View at: Google Scholar
  28. K. Khanafer and K. Vafai, “A critical synthesis of thermophysical characteristics of nanofluids,” International Journal of Heat and Mass Transfer, vol. 54, no. 19-20, pp. 4410–4428, 2011.View at: Publisher Site | Google Scholar
  29. K. Farhana, K. Kadirgama, M. M. Rahman et al., “Improvement in the performance of solar collectors with nanofluids – a state-of-the-art review,” Nano-Structures & Nano-Objects, vol. 18, p. 100276, 2019.View at: Publisher Site | Google Scholar
  30. M. Turkyilmazoglu, “Condensation of laminar film over curved vertical walls using single and two-phase nanofluid models,” European Journal of Mechanics-B/Fluids, vol. 65, pp. 184–191, 2017.View at: Publisher Site | Google Scholar
  31. X. Zhang, J. Wang, T. Wang, R. Jiang, J. Xu, and L. Zhao, “Robust feature learning for adversarial defense via hierarchical feature alignment,” Information Sciences, vol. 2020, 2020.View at: Google Scholar
  32. X. Zhang, T. Wang, W. Luo, and P. Huang, “Multi-level fusion and attention-guided CNN for image dehazing,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
  33. X. Zhang, M. Fan, D. Wang, P. Zhou, and D. Tao, “Top-k feature selection framework using robust 0-1 integer programming,” IEEE Transactions on Neural Networks and Learning Systems, vol. 1, pp. 1–15, 2020.View at: Publisher Site | Google Scholar
  34. X. Zhang, D. Wang, Z. Zhou, and Y. Ma, “Robust low-rank tensor recovery with rectification and alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 238–255, 2019.View at: Google Scholar
  35. X. Zhang, R. Jiang, T. Wang, and J. Wang, “Recursive neural network for video deblurring,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
  36. X. Zhang, T. Wang, J. Wang, G. Tang, and L. Zhao, “Pyramid channel-based feature attention network for image dehazing,” Computer Vision and Image Understanding, vol. 1, 2020.View at: Google Scholar
  37. M. Mirmozaffari, “Machine learning algorithms based on an optimization model,” 2020.View at: Google Scholar
  38. M. Mirmozaffari, M. Yazdani, A. Boskabadi, H. Ahady Dolatsara, K. Kabirifar, and N. Amiri Golilarz, “A novel machine learning approach combined with optimization models for eco-efficiency evaluation,” Applied Sciences, vol. 10, no. 15, p. 5210, 2020.View at: Publisher Site | Google Scholar
  39. M. Vosoogha and A. Addeh, “An intelligent power prediction method for wind energy generation based on optimized fuzzy system,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 5, pp. 34–43, 2019.View at: Google Scholar
  40. A. Javadi, N. Mikaeilvand, and H. Hosseinzdeh, “Presenting a new method to solve partial differential equations using a group search optimizer method (GSO),” Computational Research Progress in Applied Science and Engineering, vol. 4, no. 1, pp. 22–26, 2018.View at: Google Scholar
  41. F. J. Golrokh, Gohar Azeem, and A. Hasan, “Eco-efficiency evaluation in cement industries: DEA malmquist productivity index using optimization models,” ENG Transactions, vol. 1, pp. 1–8, 2020.View at: Google Scholar
  42. H. Yu, “Dynamic Gaussian bare-bones fruit fly optimizers with abandonment mechanism: method and analysis,” Engineering with Computers, vol. 1, pp. 1–29, 2020.View at: Google Scholar
  43. C. Yu, “SGOA: annealing-behaved grasshopper optimizer for global tasks,” Engineering with Computers, vol. 1, pp. 1–28, 2021.View at: Google Scholar
  44. W. Shan, Z. Qiao, A. A. Heidari, H. Chen, H. Turabieh, and Y. Teng, “Double adaptive weights for stabilization of moth flame optimizer: balance analysis, engineering cases, and medical diagnosis,” Knowledge-Based Systems, vol. 1, p. 106728, 2020.View at: Google Scholar
  45. J. Tu, H. Chen, J. Liu et al., “Evolutionary biogeography-based whale optimization methods with communication structure: towards measuring the balance,” Knowledge-Based Systems, vol. 212, p. 106642, 2021.View at: Publisher Site | Google Scholar
  46. Y. Zhang, “Towards augmented kernel extreme learning models for bankruptcy prediction: algorithmic behavior and comprehensive analysis,” Neurocomputing, vol. 1, 2020.View at: Google Scholar
  47. Y. Zhang, R. Liu, X. Wang, H. Chen, and C. Li, “Boosted binary Harris hawks optimizer and feature selection,” Engineering with Computers, vol. 1, pp. 1–30, 2020.View at: Google Scholar
  48. H.-L. Chen, G. Wang, C. Ma, Z.-N. Cai, W.-B. Liu, and S.-J. Wang, “An efficient hybrid kernel extreme learning machine approach for early diagnosis of Parkinson’s disease,” Neurocomputing, vol. 184, pp. 131–144, 2016.View at: Publisher Site | Google Scholar
  49. L. Hu, G. Hong, J. Ma, X. Wang, and H. Chen, “An efficient machine learning approach for diagnosis of paraquat-poisoned patients,” Computers in Biology and Medicine, vol. 59, pp. 116–124, 2015.View at: Publisher Site | Google Scholar
  50. L. Shen, H. Chen, Z. Yu et al., “Evolving support vector machines using fruit fly optimization for medical data classification,” Knowledge-Based Systems, vol. 96, pp. 61–75, 2016.View at: Publisher Site | Google Scholar
  51. J. Xia, H. Chen, Q. Li et al., “Ultrasound-based differentiation of malignant and benign thyroid Nodules: an extreme learning machine approach,” Computer Methods and Programs in Biomedicine, vol. 147, pp. 37–49, 2017.View at: Publisher Site | Google Scholar
  52. C. Li, L. Hou, B. Y. Sharma et al., “Developing a new intelligent system for the diagnosis of tuberculous pleural effusion,” Computer Methods and Programs in Biomedicine, vol. 153, pp. 211–225, 2018.View at: Publisher Site | Google Scholar
  53. X. Zhao, X. Zhang, Z. Cai et al., “Chaos enhanced grey wolf optimization wrapped ELM for diagnosis of paraquat-poisoned patients,” Computational Biology and Chemistry, vol. 78, pp. 481–490, 2019.View at: Publisher Site | Google Scholar
  54. M. Wang and H. Chen, “Chaotic multi-swarm whale optimizer boosted support vector machine for medical diagnosis,” Applied Soft Computing Journal, vol. 88, 2020.View at: Publisher Site | Google Scholar
  55. X. Xu and H.-L. Chen, “Adaptive computational chemotaxis based on field in bacterial foraging optimization,” Soft Computing, vol. 18, no. 4, pp. 797–807, 2014.View at: Publisher Site | Google Scholar
  56. R. U. Khan, X. Zhang, R. Kumar, A. Sharif, N. A. Golilarz, and M. Alazab, “An adaptive multi-layer botnet detection technique using machine learning classifiers,” Applied Sciences, vol. 9, no. 11, p. 2375, 2019.View at: Publisher Site | Google Scholar
  57. A. Addeh, A. Khormali, and N. A. Golilarz, “Control chart pattern recognition using RBF neural network with new training algorithm and practical features,” ISA Transactions, vol. 79, pp. 202–216, 2018.View at: Publisher Site | Google Scholar
  58. N. Amiri Golilarz, H. Gao, R. Kumar, L. Ali, Y. Fu, and C. Li, “Adaptive wavelet based MRI brain image de-noising,” Frontiers in Neuroscience, vol. 14, p. 728, 2020.View at: Publisher Site | Google Scholar
  59. N. A. Golilarz, H. Gao, and H. Demirel, “Satellite image de-noising with Harris hawks meta heuristic optimization algorithm and improved adaptive generalized Gaussian distribution threshold function,” IEEE Access, vol. 7, pp. 57459–57468, 2019.View at: Publisher Site | Google Scholar
  60. M. Eisazadeh and J. Rezapour, “Multi-objective optimization of the composite sheets using PSO algorithm,” 2017.View at: Google Scholar
  61. I. Bargegol, M. Nikookar, R. V. Nezafat, E. J. Lashkami, and A. M. Roshandeh, “Timing optimization of signalized intersections using shockwave theory by genetic algorithm,” Computational Research Progress in Applied Science & Engineering, vol. 1, pp. 160–167, 2015.View at: Google Scholar
  62. B. Bai, Z. Guo, C. Zhou, W. Zhang, and J. Zhang, “Application of adaptive reliability importance sampling-based extended domain PSO on single mode failure in reliability engineering,” Information Sciences, vol. 546, pp. 42–59, 2021.View at: Publisher Site | Google Scholar
  63. J. Liu, C. Wu, G. Wu, and X. Wang, “A novel differential search algorithm and applications for structure design,” Applied Mathematics and Computation, vol. 268, pp. 246–269, 2015.View at: Publisher Site | Google Scholar
  64. X. Zhao, D. Li, B. Yang, C. Ma, Y. Zhu, and H. Chen, “Feature selection based on improved ant colony optimization for online detection of foreign fiber in cotton,” Applied Soft Computing, vol. 24, pp. 585–596, 2014.View at: Publisher Site | Google Scholar
  65. D. Zhao, “Chaotic random spare ant colony optimization for multi-threshold image segmentation of 2D Kapur entropy,” Knowledge-Based Systems, vol. 24, p. 106510, 2020.View at: Google Scholar
  66. H. Chen, A. A. Heidari, H. Chen, M. Wang, Z. Pan, and A. H. Gandomi, “Multi-population differential evolution-assisted Harris hawks optimization: framework and case studies,” Future Generation Computer Systems, vol. 111, pp. 175–198, 2020.View at: Publisher Site | Google Scholar
  67. J. Hu, H. Chen, A. A. Heidari et al., “Orthogonal learning covariance matrix for defects of grey wolf optimizer: insights, balance, diversity, and feature selection,” Knowledge-Based Systems, vol. 213, p. 106684, 2021.View at: Publisher Site | Google Scholar
  68. G. Sun, B. Yang, Z. Yang, and G. Xu, “An adaptive differential evolution with combined strategy for global numerical optimization,” Soft Computing, vol. 24, pp. 1–20, 2019.View at: Google Scholar
  69. G. Sun, C. Li, and L. Deng, “An adaptive regeneration framework based on search space adjustment for differential evolution,” Neural Computing and Applications, vol. 24, pp. 1–17, 2021.View at: Google Scholar
  70. A. Addeh and M. Iri, “Brain tumor type classification using deep features of MRI images and optimized RBFNN,” ENG Transactions, vol. 2, pp. 1–7, 2021.View at: Google Scholar
  71. F. J. Golrokh and A. Hasan, “A comparison of machine learning clustering algorithms based on the DEA optimization approach for pharmaceutical companies in developing countries,” Soft Computing, vol. 1, pp. 1–8, 2020.View at: Google Scholar
  72. H. Tyagi, P. Phelan, and R. Prasher, “Predicted efficiency of a low-temperature nanofluid-based direct absorption solar collector,” Journal of Solar Energy Engineering, vol. 131, no. 4, 2009.View at: Publisher Site | Google Scholar
  73. S. Rashidi, M. Bovand, and J. A. Esfahani, “Heat transfer enhancement and pressure drop penalty in porous solar heat exchangers: a sensitivity analysis,” Energy Conversion and Management, vol. 103, pp. 726–738, 2015.View at: Publisher Site | Google Scholar
  74. N. Akram, R. Sadri, S. N. Kazi et al., “A comprehensive review on nanofluid operated solar flat plate collectors,” Journal of Thermal Analysis and Calorimetry, vol. 139, no. 2, pp. 1309–1343, 2020.View at: Publisher Site | Google Scholar
Fig. 11. Velocity vectors along x-direction through the center of the box culvert for B0, B30, B50, and B70 respectively.

Numerical investigation of scour characteristics downstream of blocked culverts

막힌 암거 하류의 세굴 특성 수치 조사

NesreenTahabMaged M.El-FekyaAtef A.El-SaiadaIsmailFathya
aDepartment of Water and Water Structures Engineering, Faculty of Engineering, Zagazig University, Zagazig 44519, Egypt
bLab Manager, Faculty of Engineering, Zagazig University, Zagazig 44519, Egypt

Abstract

횡단 구조물을 통한 막힘은 안정성을 위협하는 위험한 문제 중 하나입니다. 암거의 막힘 형상 및 하류 세굴 특성에 미치는 영향에 관한 연구는 거의 없습니다.

이 연구의 목적은 수면과 세굴 모두에서 상자 암거를 통한 막힘의 작용을 수치적으로 논의하는 것입니다. 이를 위해 FLOW 3D v11.1.0을 사용하여 퇴적물 수송 모델을 조사했습니다.

상자 암거를 통한 다양한 차단 비율이 연구되었습니다. FLOW 3D 모델은 실험 데이터로 보정되었습니다. 결과는 FLOW 3D 프로그램이 세굴 다운스트림 상자 암거를 정확하게 시뮬레이션할 수 있음을 나타냅니다.

막힌 경우에 대한 속도 분포, 최대 세굴 깊이 및 수심을 플롯하고 비차단된 사례(기본 사례)와 비교했습니다.

그 결과 암거 높이의 70% 차단율은 상류의 수심을 암거 높이의 2.3배 증가시키고 평균 유속은 기본 경우보다 3배 더 증가시키는 것으로 입증되었다. 막힘 비율의 함수로 상대 최대 세굴 깊이를 추정하는 방정식이 만들어졌습니다.

Blockage through crossing structures is one of the dangerous problems that threaten its stability. There are few researches concerned with blockage shape in culverts and its effect on characteristics of scour downstream it.

The study’s purpose is to discuss the action of blockage through box culvert on both water surface and scour numerically. A sediment transport model has been investigated for this purpose using FLOW 3D v11.1.0. Different ratios of blockage through box culvert have been studied. The FLOW 3D model was calibrated with experimental data.

The results present that the FLOW 3D program was capable to simulate accurately the scour downstream box culvert. The velocity distribution, maximum scour depth and water depths for blocked cases have been plotted and compared with the non-blocked case (base case).

The results proved that the blockage ratio 70% of culvert height makes the water depth upstream increases by 2.3 times of culvert height and mean velocity increases by 3 times more than in the base case. An equation has been created to estimate the relative maximum scour depth as a function of blockage ratio.

1. Introduction

Local scour is the removal of granular bed material by the action of hydrodynamic forces. As the depth of scour hole increases, the stability of the foundation of the structure may be endangered, with a consequent risk of damage and failure [1]. So the prediction and control of scour is considered to be very important for protecting the water structures from failure. Most previous studies were designed to study the different factors that impact on scour and their relationship with scour hole dimensions like fluid characteristics, flow conditions, bed properties, and culvert geometry. Many previous researches studied the effect of flow rate on scour hole by information Froude number or modified Froude number [2][3][4][5][6]. Cesar Mendoza [6] found a good correlation between the scour depth and the discharge Intensity (Qg−.5D−2.5). Breusers and Raudkiv [7] used shear velocity in the outlet-scour prediction procedure. Ali and Lim [8] used the densimetric Froude number in estimation of the scour depth [1][8][9][10][11][12][13][14]. “The densimetric Froude number presents the ratio of the tractive force on sediment particle to the submerged specific weight of the sediment” [15](1)Fd=uρsρ-1gD50

Ali and Lim [8] pointed to the consequence of tailwater depth on scour behavior [1][2][8][13]. Abida and Townsend [2] indicated that the maximum depth of local scour downstream culvert was varying with the tailwater depth in three ways: first, for very shallow tailwater depths, local scouring decreases with a decrease in tailwater depth; second, when the ratio of tailwater depth to culvert height ranged between 0.2 and 0.7, the scour depth increases with decreasing tailwater depth; and third for a submerged outlet condition. The tailwater depth has only a marginal effect on the maximum depth of scour [2]. Ruff et al. [16] observed that for materials having similar mean grain sizes (d50) but different standard deviations (σ). As (σ) increased, the maximum scour hole depth decreased. Abt et al. [4] mentioned to role of soil type of maximum scour depth. It was noticed that local scour was more dangerous for uniform sands than for well-graded mixtures [1][2][4][9][17][18]. Abt et al [3][19] studied the culvert shape effect on scour hole. The results evidenced that the culvert shape has a limited effect on outlet scour. Under equivalent discharge conditions, it was noted that a square culvert with height equal to the diameter of a circular culvert would reduce scour [16][20]. The scour hole dimension was also effected by the culvert slope. Abt et al. [3][21] showed that the culvert slope is a key element in estimating the culvert flow velocity, the discharge capacity, and sediment transport capability. Abt et al. [21][22] tested experimentally culvert drop height effect on maximum scour depth. It was observed that as the drop height was increasing, the depth of scour was also increasing. From the previous studies, it could have noticed that the most scour prediction formula downstream unblocked culvert was the function of densimetric Froude number, soil properties (d50, σ), tailwater depth and culvert opening size. Blockage is the phenomenon of plugging water structures due to the movement of water flow loaded with sediment and debris. Water structures blockage has a bad effect on water flow where it causes increasing of upstream water level that may cause flooding around the structure and increase of scour rate downstream structures [23][24]. The blockage phenomenon through was studied experimentally and numerical [15][25][26][27][28][29][30][31][32][33]. Jaeger and Lucke [33] studied the debris transport behavior in a natural channel in Australia. Froude number scale model of an existing culvert was used. It was noticed that through rainfall event, the mobility of debris was impressed by stream shape (depth and width). The condition of the vegetation (size and quantities) through the catchment area was the main factor in debris transport. Rigby et al. [26] reported that steep slope was increasing the ability to mobilize debris that form field data of blocked culverts and bridges during a storm in Wollongong city.

Streftaris et al. [32] studied the probability of screen blockage by debris at trash screens through a numerical model to relate between the blockage probability and nature of the area around. Recently, many commercial computational fluid programs (CFD) such as SSIIM, Fluent, and FLOW 3D are used in the analysis of the scour process. Scour and sediment transport numerical model need to validate by using experimental data or field data [34][35][36][37][38]. Epely-Chauvin et al. [36] investigated numerically the effect of a series of parallel spur diked. The experimental data were compared by SSIIM and FLOW 3D program. It was found that the accuracy of calibrated FLOW 3D model was better than SSIIM model. Nielsen et al. [35] used the physical model and FLOW 3D model to analyze the scour process around the pile. The soil around the pile was uniform coarse stones in the physical models that were simulated by regular spheres, porous media, and a mixture of them. The calibrated porous media model can be used to determine the bed shear stress. In partially blocked culverts, there aren’t many studies that explain the blockage impact on scour dimensions. Sorourian et al. [14][15] studied the effect of inlet partial blockage on scour characteristics downstream box culvert. It resulted that the partial blockage at the culvert inlet could be the main factor in estimating the depth of scour. So, this study is aiming to investigate the effects of blockage through a box culvert on flow and scour characteristics by different blockage ratios and compares the results with a non-blocked case. Create a dimensionless equation relates the blockage ratio of the culvert with scour characteristics downstream culvert.

2. Experimental data

The experimental work of the study was conducted in the Hydraulics and Water Engineering Laboratory, Faculty of Engineering, Zagazig University, Egypt. The flume had a rectangular cross-section of 66 cm width, 65.5 cm depth, and 16.2 m long. A rectangular culvert was built with 0.2 m width, 0.2 m height and 3.00 m long with θ = 25° gradually outlet and 0.8 m fixed apron. The model was located on the mid-point of the channel. The sediment part was extended for a distance 2.20 m with 0.66 m width and 0.20 m depth of coarse sand with specific weight 1.60 kg/cm3, d50 = 2.75 mm and σ (d90/d50) = 1.50. The particle size distribution was as shown in Fig. 1. The experimental model was tested for different inlet flow (Q) of 25, 30, 34, 40 l/s for different submerged ratio (S) of 1.25, 1.50, 1.75.

3. Dimensional analysis

A dimensional analysis has been used to reduce the number of variables which affecting on the scour pattern downstream partial blocked culvert. The main factors affecting the maximum scour depth are:(2)ds=f(b.h.L.hb.lb.Q.ud.hu.hd.D50.ρ.ρs.g.ls.dd.ld)

Fig. 2 shows a definition sketch of the experimental model. The maximum scour depth can be written in a dimensionless form as:(3)dsh=f(B.Fd.S)where the ds/h is the relative maximum scour depth.

4. Numerical work

The FLOW 3D is (CFD) program used by many researchers and appeared high accuracy in solving hydrodynamic and sediment transport models in the three dimensions. Numerical simulation with FLOW 3D was performed to study the impacts of blockage ratio through box culvert on shear stress, velocity distribution and the sediment transport in terms of the hydrodynamic features (water surface, velocity and shear stress) and morphological parameters (scour depth and sizes) conditions in accurately and efficiently. The renormalization group (RNG) turbulence model was selected due to its high ability to predict the velocity profiles and turbulent kinetic energy for the flow through culvert [39]. The one-fluid incompressible mode was used to simulate the water surface. Volume of fluid (VOF) method was employed in FLOW 3D to tracks a liquid interface through arbitrary deformations and apply the correct boundary conditions at the interface [40].1.

Governing equations

Three-dimensional Reynolds-averaged Navier Stokes (RANS) equation was applied for incompressible viscous fluid motion. The continuity equation is as following:(4)VF∂ρ∂t+∂∂xρuAx+∂∂yρvAy+∂∂zρwAz=RDIF(5)∂u∂t+1VFuAx∂u∂x+vAy∂u∂y+ωAz∂u∂z=-1ρ∂P∂x+Gx+fx(6)∂v∂t+1VFuAx∂v∂x+vAy∂v∂y+ωAz∂v∂z=-1ρ∂P∂y+Gy+fy(7)∂ω∂t+1VFuAx∂ω∂x+vAy∂ω∂y+ωAz∂ω∂z=-1ρ∂P∂z+Gz+fz

ρ is the fluid density,

VF is the volume fraction,

(x,y,z) is the Cartesian coordinates,

(u,v,w) are the velocity components,

(Ax,Ay,Az) are the area fractions and

RDIF is the turbulent diffusion.

P is the average hydrodynamic pressure,

(Gx, Gy, Gz) are the body accelerations and

(fx, fy, fz) are the viscous accelerations.

The motion of sediment transport (suspended, settling, entrainment, bed load) is estimated by predicting the erosion, advection and deposition process as presented in [41].

The critical shields parameter is (θcr) is defined as the critical shear stress τcr at which sediments begin to move on a flat and horizontal bed [41]:(8)θcr=τcrgd50(ρs-ρ)

The Soulsby–Whitehouse [42] is used to predict the critical shields parameter as:(9)θcr=0.31+1.2d∗+0.0551-e(-0.02d∗)(10)d∗=d50g(Gs-1ν3where:

d* is the dimensionless grain size

Gs is specific weight (Gs = ρs/ρ)

The entrainment coefficient (0.005) was used to scale the scour rates and fit the experimental data. The settling velocity controls the Soulsby deposition equation. The volumetric sediment transport rate per width of the bed is calculated using Van Rijn [43].2.

Meshing and geometry of model

After many trials, it was found that the uniform cell size with 0.03 m cell size is the closest to the experimental results and takes less time. As shown in Fig. 3. In x-direction, the total model length in this direction is 700 cm with mesh planes at −100, 0, 300, 380 and 600 cm respectively from the origin point, in y-direction, the total model length in this direction is 66 cm at distances 0, 23, 43 and 66 cm respectively from the origin point. In z-direction, the total model length in this direction is 120 cm. with mesh planes at −20, 0, 20 and 100 cm respectively.3.

Boundary condition

As shown in Fig. 4, the boundary conditions of the model have been defined to simulate the experimental flow conditions accurately. The upstream boundary was defined as the volume flow rate with a different flow rate. The downstream boundary was defined as specific pressure with different fluid elevation. Both of the right side, the left side, and the bottom boundary were defined as a wall. The top boundary defined as specified pressure with pressure value equals zero.

5. Validation of experimental results and numerical results

The experimental results investigated the flow and scour characteristics downstream culvert due to different flow conditions. The measured value of maximum scour depth is compared with the simulated depth from FLOW 3D model as shown in Fig. 5. The scour results show that the simulated results from the numerical model is quite close to the experimental results with an average error of 3.6%. The water depths in numerical model results is so close to the experimental results as shown in Fig. 6 where the experiment and numerical results are compared at different submerged ratios and flow rates. The results appear maximum error percentage in water depths upstream and downstream the culvert is about 2.37%. This indicated that the FLOW 3D is efficient for the prediction of maximum scour depth and the flow depths downstream box culvert.

6. Computation time

The run time was chosen according to reaching to the stability limit. Hydraulic stability was achieved after 50 s, where the scour development may still go on. For run 1, the numerical simulation was run for 1000 s as shown in Fig. 7 where it mostly reached to scour stability at 800 s. The simulation time was taken 500 s at about 95% of scour stability.

7. Analysis and discussions

Fig. 8 shows the study sections where sec 1 represents to upstream section, sec2 represents to inside section and sec3 represents to downstream stream section. Table 1 indicates the scour hole dimensions at different blockage case. The symbol (B) represents to blockage and the number points to blockage ratio. B0 case signifies to the non-blocked case, B30 is that blockage height is 30% to the culvert height and so on.

Table 1. The scour results of different blockage ratio.

Casehb cmB = hb/hQ lit/sSFdd50 mmds/h measuredls/hdd/hld/hds/h estimated
B000351.261.692.50.581.500.275.000.46
B3060.30351.261.682.50.481.250.274.250.40
B50100.50351.221.742.50.451.100.244.000.37
B70140.70351.231.732.50.431.500.165.500.33

7.1. Scour hole geometry

The scour hole geometry mainly depends on the properties of soil of the bed downstream the fixed apron. From Table 1, the results show that the maximum scour depth in B0 case is about 0.58 of culvert height while the maximum deposition in B0 is 0.27 culvert height. There is a symmetric scour hole as shown in Fig. 9 in B0 case. An asymmetric scour hole is created in B50 and B70 due to turbulences that causes the deviation of the jet direction from the center of the flume where appear in Fig. 11 and Fig. 19.

7.2. Flow water surface

Fig. 10 presents the relative free surface water (hw/h) along the x-direction at center of the box culvert. From the mention Figure, it is easy to release the effect of different blockage ratios. The upstream water level rises by increasing the blockage ratio. Increasing upstream water level may cause flooding over the banks of the waterway. In the 70% blockage case, the upstream water level rises to 2.3 times of culvert height more than the non-blocked case at the same discharge and submerged ratio. The water surface profile shows an increase in water level upstream the culvert due to a decrease in transverse velocity. Because of decreasing velocity downstream culvert, there is an increase in water level before it reaches its uniform depth.

7.3. Velocity vectors

Scour downstream hydraulic structures mainly affects by velocities distribution and bed shear stress. Fig. 11 shows the velocity vectors and their magnitude in xz plane at the same flow conditions. The difference in the upstream water level due to the different blockage ratios is so clear. The maximum water level is in B70 and the minimum level is in B0. The inlet mean velocity value is about 0.88 m/s in B0 increases to 2.86 m/s in B70. As the blockage ratio increases, the inlet velocity increases. The outlet velocity in B0 case makes downward jet causes scour hole just after the fixed apron in the middle of the bed while the blockage causes upward water flow that appears clearly in B70. The upward jet decreases the scour depth to 0.13 culvert height less than B0 case. After the scour hole, the velocity decreases and the flow becomes uniform.

7.4. Velocity distribution

Fig. 12 represents flow velocity (Vx) distribution along the vertical depth (z/hu) upstream the inlet for the different blockage ratios at the same flow conditions. From the Figure, the maximum velocity creates closed to bed in B0 while in blocked case, the maximum horizontal velocity creates at 0.30 of relative vertical depth (z/hu). Fig. 13 shows the (Vz) distribution along the vertical depth (z/hu) upstream culvert at sec 1. From the mentioned Figure, it is easy to note that the maximum vertical is in B70 which appears that as the blockage ratio increases the vertical ratio also increases. In the non-blocked case. The vertical velocity (Vz) is maximum at (z/hu) equals 0.64. At the end of the fixed apron (sec 3), the horizontal velocity (Vx) is slowly increasing to reach the maximum value closed to bed in B0 and B30 while the maximum horizontal velocity occurs near to the top surface in B50 and B70 as shown in Fig. 14. The vertical velocity component along the vertical depth (z/hd) is presented in Fig. 15. The vertical velocity (Vz) is maximum in B0 at vertical depth (z/hd) 0.3 with value 0.45 m/s downward. Figs. 16 and 17 observe velocity components (Vx, Vz) along the vertical depth just after the end of blockage length at the centerline of the culvert barrel. It could be noticed the uniform velocity distribution in B0 case with horizontal velocity (Vx) closed to 1.0 m/s and vertical velocity closed to zero. In the blocked case, the maximum horizontal velocity occurs in depth more than the blockage height.

7.5. Bed velocity distribution

Fig. 18 presents the x-velocity vectors at 1.5 cm above the bed for different blockage ratios from the velocity vectors distribution and magnitude, it is easy to realize the position of the scour hole and deposition region. In B0 and B30, the flow is symmetric so that the scour hole is created around the centerline of flow while in B50 and B70 cases, the flow is asymmetric and the scour hole creates in the right of flow direction in B50. The maximum scour depth is found in the left of flow direction in B70 case where the high velocity region is found.

8. Maximum scour depth prediction

Regression analysis is used to estimate maximum scour depth downstream box culvert for different ratios of blockage by correlating the maximum relative scour by other variables that affect on it in one formula. An equation is developed to predict maximum scour depth for blocked and non-blocked. As shown in the equation below, the relative maximum scour depth(ds/hd) is a function of densimetric Froude number (Fd), blockage ratio (B) and submerged ratio (S)(11)dsh=0.56Fd-0.20B+0.45S-1.05

In this equation the coefficient of correlation (R2) is 0.82 with standard error equals 0·08. The developed equation is valid for Fd = [0.9 to 2.10] and submerged ratio (S) ≥ 1.00. Fig. 19 shows the comparison between relative maximum scour depths (ds/h) measured and estimated for different blockage ratios. Fig. 20 clears the comparison between residuals and ds/h estimated for the present study. From these figures, it could be noticed that there is a good agreement between the measured and estimated relative scour depth.

9. Comparison with previous scour equations

Many previous scour formulae have been produced for calculation the maximum scour depth downstream non-blockage culvert. These equations have been included the effect of flow regime, culvert shape, soil properties and the flow rate on maximum scour depth. Two of previous experimental studies data have been chosen to be compared with the present study results in non-blocked study data. Table 2 shows comparison of culvert shape, densmetric Froude number, median particle size and scour equations for these previous studies. By applying the present study data in these studies scour formula as shown in Fig. 21, it could be noticed that there are a good agreement between present formula results and others empirical equations results. Where that Lim [44] and Abt [4] are so closed to the present study data.

Table 2. Comparison of some previous scour formula.

ResearchersFdCulvert shaped50(mm)Proposed equationSubmerged ratio
Present study0.9–2.11square2.75dsh=0.56Fd-0.20B+0.45S-1.051.25–1.75
Lim [44]1–10Circular1.65dsh=0.45Fd0.47
Abt [4]Fd ≥ 1Circular0.22–7.34-dsh=3.67Fd0.57∗D500.4∗σ-0.4

10. Conclusions

The present study has shown that the FLOW 3D model can accurately simulate water surface and the scour hole characteristics downstream the box culvert with error percentage in water depths does not exceed 2.37%. Velocities distribution through and outlets culvert barrel helped on understanding the scour hole shape.

The blockage through culvert had caused of increasing of water surface upstream structure where the upstream water level in B70 was 2.3 of culvert height more than non-blocked case at the same discharge that could be dangerous on the stability of roads above. The depth averaged velocity through culvert barrel increased by 3 times its value in non-blocked case.

On the other hand, blockage through culvert had a limited effect on the maximum scour depth. The little effect of blockage on maximum scour depth could be noticed in Fig. 11. From this Figure, it could be noted that the residual part of culvert barrel after the blockage part had made turbulences. These turbulences caused the deviation of the flow resulting in the formation of asymmetric scour hole on the side of channel. This not only but in B70 the blockage height caused upward jet which made a wide far scour hole as cleared from the results in Table 1.

An empirical equation was developed from the results to estimate the maximum scour depth relative to culvert height function of blockage ratio (B), submerged ratio (S), and densimetric Froude number (Fd). The equation results was compared with some scour formulas at the same densimetric Froude number rang where the present study results was in between the other equations results as shown in Fig. 21.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

[1]P. Sarathi, M. Faruque, R. BalachandarInfluence of tailwater depth, sediment size and densimetric Froude number on scour by submerged square wall jetsJ. Hydraul. Res., 46 (2) (2008), pp. 158-175CrossRefView Record in ScopusGoogle Scholar[2]H. Abida, R. TownsendLocal scour downstream of box-culvert outletsJ. Irrig. Drain. Eng., 117 (3) (1991), pp. 425-440CrossRefView Record in ScopusGoogle Scholar[3]S.R. Abt, C.A. Donnell, J.F. Ruff, F.K. DoehringCulvert Slope and Shape Effects on Outlet ScourTransp. Res. Rec., 1017 (1985), pp. 24-30View Record in ScopusGoogle Scholar[4]S.R. Abt, R.L. Kloberdanz, C. MendozaUnified culvert scour determinationJ. Hydraul. Eng., 110 (10) (1984), pp. 1475-1479CrossRefView Record in ScopusGoogle Scholar[5]J.P. Bohan, Erosion And Riprap Requirements At Culvert And Storm-Drain Outlets, ARMY ENGINEER WATERWAYS EXPERIMENT STATION VICKSBURG MISS1970.Google Scholar[6]C. Mendoza, S.R. Abt, J.F. RuffHeadwall influence on scour at culvert outletsJ. Hydraul. Eng., 109 (7) (1983), pp. 1056-1060CrossRefView Record in ScopusGoogle Scholar[7]H. Breusers, A. Raudkivi, Scouring, hydraulic structures design manual, vol. 143, IAHR, AA Balkema, Rotterdam, 1991.Google Scholar[8]K. Ali, S. LimLocal scour caused by submerged wall jetsProc. Inst. Civ. Eng., 81 (4) (1986), pp. 607-645CrossRefView Record in ScopusGoogle Scholar[9]O. Aderibigbe, N. RajaratnamEffect of sediment gradation on erosion by plane turbulent wall jetsJ. Hydraul. Eng., 124 (10) (1998), pp. 1034-1042View Record in ScopusGoogle Scholar[10]F.W. Blaisdell, C.L. AndersonA comprehensive generalized study of scour at cantilevered pipe outletsJ. Hydraul. Res., 26 (4) (1988), pp. 357-376CrossRefView Record in ScopusGoogle Scholar[11]Y.-M. Chiew, S.-Y. LimLocal scour by a deeply submerged horizontal circular jetJ. Hydraul. Eng., 122 (9) (1996), pp. 529-532View Record in ScopusGoogle Scholar[12]R.A. Day, S.L. Liriano, W.R. WhiteEffect of tailwater depth and model scale on scour at culvert outletsProc. Instit. Civil Eng. – Water Marit. Eng., 148 (3) (2001), pp. 189-198http://www.icevirtuallibrary.com/doi/10.1680/wame.2001.148.3.18910.1680/wame.2001.148.3.189View Record in ScopusGoogle Scholar[13]S. Emami, A.J. SchleissPrediction of localized scour hole on natural mobile bed at culvert outletsScour and Erosion (2010), pp. 844-853CrossRefView Record in ScopusGoogle Scholar[14]S. Sorourian, A. Keshavarzi, J. Ball, B. SamaliStudy of Blockage Effect on Scouring Pattern Downstream of a Box Culvert under Unsteady FlowAustr. J Water Resor. (2013)Google Scholar[15]S. Sorourian, Turbulent Flow Characteristics At The Outlet Of Partially Blocked Box Culverts, in: 36th IAHR World Congress, The Hague, the Netherlands, 2015.Google Scholar[16]J. Ruff, S. Abt, C. Mendoza, A. Shaikh, R. KloberdanzScour at culvert outlets in mixed bed materialsUnited States. Federal Highway Administration. Office of Research and Development (1982)Google Scholar[17]S.A. Ansari, U.C. Kothyari, K.G.R. RajuInfluence of cohesion on scour under submerged circular vertical jetsJ. Hydraul. Eng., 129 (12) (2003), pp. 1014-1019View Record in ScopusGoogle Scholar[18]B. Crookston B. Tullis, Scour and Riprap Protection in a Bottomless Arch Culvert, in: World Environmental and Water Resources Congress 2008: Ahupua’A, 2008, pp. 1–10.Google Scholar[19]S.R. Abt, J. Ruff, F. Doehring, C. DonnellInfluence of culvert shape on outlet scourJ. Hydraul. Eng., 113 (3) (1987), pp. 393-400View Record in ScopusGoogle Scholar[20]Y.H. Chen, Scour at outlets of box culverts, Colorado State University, 1970.Google Scholar[21]S. Abt, P. Thompson, T. LewisEnhancement of the culvert outlet scour estimation equationsTransp. Res. Rec. J. Transp. Res. Board, 1523 (1996), pp. 178-185View Record in ScopusGoogle Scholar[22]F.K. Doehring, S.R. AbtDrop height influence on outlet scourJ. Hydraul. Eng., 120 (12) (1994), pp. 1470-1476CrossRefView Record in ScopusGoogle Scholar[23]W. Weeks, A. Barthelmess, E. Rigby, G. Witheridge, R. Adamson, Australian rainfall and runoff revison project 11: blockage of hydraulic structures, 2009.Google Scholar[24]W. Weeks, G. Witheridge, E. Rigby, A. BarthelmessProject 11: blockage of hydraulic structuresEngineers Australia (2013)Google Scholar[25]S.R. Abt, T.E. Brisbane, D.M. Frick, C.A. McKnightTrash rack blockage in supercritical flowJ. Hydraul. Eng., 118 (12) (1992), pp. 1692-1696View Record in ScopusGoogle Scholar[26]E. Rigby, M. Boyd, S. Roso, P. Silveri, A. Davis, Causes and effects of culvert blockage during large storms, in: Global solutions for urban drainage, 2002, pp. 1–16.Google Scholar[27]S. Roso, M. Boyd, E. Rigby, R. VanDrie“Prediction of increased flooding in urban catchments due to debris blockage and flow diversionsProceedings Novatech (2004), pp. 8-13View Record in ScopusGoogle Scholar[28]C.-D. Jan, C.-L. ChenDebris flows caused by Typhoon Herb in Taiwanin Debris-Flow Hazards and Related Phenomena, Springer (2005), pp. 539-563CrossRefGoogle Scholar[29]L.W. Zevenbergen, P.F. Lagasse, P.E. Clopper, Effects of debris on bridge pier scour, in: World Environmental and Water Resources Congress 2007: Restoring Our Natural Habitat, 2007, pp. 1–10.Google Scholar[30]A. Barthelmess, E. Rigby, Estimating Culvert and Bridge Blockages-a Simplified Procedure, in: Proceedings of the 34th World Congress of the International Association for Hydro-Environment Research and Engineering: 33rd Hydrology and Water Resources Symposium and 10th Conference on Hydraulics in Water Engineering, Engineers Australia, 2011, pp. 39.Google Scholar[31]E. Rigby, A. Barthelmess, Culvert Blockage Mechanisms and their Impact on Flood Behaviour, in: Proceedings of the 34th World Congress of the International Association for Hydro-Environment Research and Engineering: 33rd Hydrology and Water Resources Symposium and 10th Conference on Hydraulics in Water Engineering, Engineers Australia, 2011, pp. 380.Google Scholar[32]G. Streftaris, N. Wallerstein, G. Gibson, S. ArthurModeling probability of blockage at culvert trash screens using Bayesian approachJ. Hydraul. Eng., 139 (7) (2012), pp. 716-726Google Scholar[33]R. Jaeger, T. LuckeInvestigating the relationship between rainfall intensity, catchment vegetation and debris mobilityInt. J. GEOMATE, 12 (33) (2017), pp. 22-29 Download PDFView Record in ScopusGoogle Scholar[34]S. Amiraslani, J. Fahimi, H. Mehdinezhad, The Numerical Investigation of Free Falling Jet’s Effect On the Scour of Plunge Pool, in: XVIII International conference on water resources, Tehran University, Iran, 2008.Google Scholar[35]A.W. Nielsen, X. Liu, B.M. Sumer, J. FredsøeFlow and bed shear stresses in scour protections around a pile in a currentCoast. Eng., 72 (2013), pp. 20-38ArticleDownload PDFView Record in ScopusGoogle Scholar[36]G. Epely-Chauvin, G. De Cesare, S. SchwindtNumerical modelling of plunge pool scour evolution in non-cohesive sedimentsEng. Appl. Comput. Fluid Mech., 8 (4) (2014), pp. 477-487 Download PDFCrossRefView Record in ScopusGoogle Scholar[37]H. Karami, H. Basser, A. Ardeshir, S.H. HosseiniVerification of numerical study of scour around spur dikes using experimental dataWater Environ. J., 28 (1) (2014), pp. 124-134CrossRefView Record in ScopusGoogle Scholar[38]S.-H. Oh, K.S. Lee, W.-M. JeongThree-dimensional experiment and numerical simulation of the discharge performance of sluice passageway for tidal power plantRenew. Energy, 92 (2016), pp. 462-473ArticleDownload PDFView Record in ScopusGoogle Scholar[39]M.A. Khodier, B.P. TullisExperimental and computational comparison of baffled-culvert hydrodynamics for fish passageJ. Appl. Water Eng. Res. (2017), pp. 1-9CrossRefView Record in ScopusGoogle Scholar[40]F.S. Inc., FLOW-3D user’s manual, Flow Science, Inc., 2009.Google Scholar[41]G. Wei, J. Brethour, M. Grünzner, J. BurnhamSedimentation scour modelFlow Science Report, 7 (2014), pp. 1-29View Record in ScopusGoogle Scholar[42]R. Soulsby, R. Whitehouse, Threshold of sediment motion in coastal environments, in: Pacific Coasts and Ports’ 97: Proceedings of the 13th Australasian Coastal and Ocean Engineering Conference and the 6th Australasian Port and Harbour Conference, vol. 1, Centre for Advanced Engineering, University of Canterbury, 1997, pp. 145.Google Scholar[43]L.C.v. RijnSediment transport, part II: suspended load transportJ. Hydraul. Eng., 110 (11) (1984), pp. 1613-1641View Record in ScopusGoogle Scholar[44]S Y LIMScour below unsubmerged full-flowing culvert outletsProc. Instit. Civil Eng. – Water Marit. Energy, 112 (2) (1995), pp. 136-149http://www.icevirtuallibrary.com/doi/10.1680/iwtme.1995.2765910.1680/iwtme.1995.27659View Record in ScopusGoogle Scholar

Peer review under responsibility of Faculty of Engineering, Alexandria University.

Figure 1- Schematic diagram of pooled stepped spillway conducted by Felder et al. (2012A): Notes: h step height (10 cm): w pool height (3.1 cm): l horizontal step length (20 cm): lw pool weir length (1.5 cm): d' is the water depth above the crest; y' is the distance normal to the crest invert

Study of inception point, void fraction and pressure over pooled stepped spillways using Flow-3D

Khosro Morovati , Afshin Eghbalzadeh 
International Journal of Numerical Methods for Heat & Fluid Flow

ISSN: 0961-5539

Article publication date: 3 April 2018

Abstract

많은 계단식 배수로 지오메트리 설계 지침이 평평한 단계를 위해 개발되었지만 통합 단계를 설계하는 것이 더 효율적으로 작동하는 배수로에 대한 적절한 대안이 될 수 있습니다.

이 논문은 POOL의 다른 높이에서 공기 연행과 보이드 비율의 시작점을 다루는 것을 목표로 합니다. 그 후, FLOW-3D 소프트웨어를 사용하여 POOL과 경사면의 높이를 다르게 하여 폭기된 지역과 폭기되지 않은 지역에서 압력 분포를 평가했습니다.

얻어진 수치 결과와 실험 결과의 비교는 본 연구에 사용된 모든 방류에 대해 잘 일치했습니다. POOL 높이는 시작 지점 위치에 미미한 영향을 미쳤습니다. 공극률의 값은 높은 방류에 비해 낮은 방전에서 더 많은 영향을 받았습니다.

여수로의 마루(통기되지 않은 지역)에서는 음압이 나타나지 않았으며 각 방류에서 마루를 따라 높이가 15cm인 수영장에서 최대 압력 값이 얻어졌습니다.

모든 사면에서 웅덩이 및 평평한 계단형 여수로의 계단층 부근에서는 음압이 형성되지 않았습니다. 그러나 평단식 여수로에 비해 평단식 여수로의 수직면 부근에서 음압이 더 많이 형성되어 평단식 슈트에서 캐비테이션 현상이 발생할 확률이 증가하였습니다.

Study of inception point, void fraction and pressure over pooled
stWhile many stepped spillways geometry design guidelines were developed for flat steps, designing pooled steps might be an appropriate alternative to spillways working more efficiency. This paper aims to deal with the inception point of air-entrainment and void fraction in the different height of the pools. Following that, pressure distribution was evaluated in aerated and non-aerated regions under the effect of different heights of the pools and slopes through the use of the FLOW-3D software. Comparison of obtained numerical results with experimental ones was in good agreement for all discharges used in this study. Pools height had the insignificant effect on the inception point location. The value of void fraction was more affected in lower discharges in comparison with higher ones. Negative pressure was not seen over the crest of spillway (non-aerated region), and the maximum pressure values were obtained for pools with 15 cm height along the crest in each discharge. In all slopes, negative pressure was not formed near the step bed in the pooled and flat stepped spillways. However, negative pressure was formed in more area near the vertical face in the flat stepped spillway compared with the pooled stepped spillway which increases the probability of cavitation phenomenon in the flat stepped chute.

Design/methodology/approach

압력, 공극률 및 시작점을 평가하기 위해 POOL된 계단식 여수로가 사용되었습니다. 또한 POOL의 다른 높이가 사용되었습니다. 이 연구의 수치 시뮬레이션은 Flow-3D 소프트웨어를 통해 수행되었습니다. 얻어진 결과는 풀이 압력, 공극률 및 시작점을 포함한 2상 유동 특성에 영향을 미칠 수 있음을 나타냅니다.

Findings

마루 위에는 음압이 보이지 않았습니다. 압력 값은 사용된 모든 높이와 15cm 높이에서 얻은 최대 값에 대해 다릅니다. 또한, 풀링 스텝은 플랫 케이스에 비해 음압점 감소에 더 효과적인 역할을 하였습니다. 시작 지점 위치는 특히 9 및 15cm 높이에 대해 스키밍 흐름 영역과 비교하여 낮잠 및 전환 흐름 영역에서 더 많은 영향을 받았습니다.

Keywords

Citation

Morovati, K. and Eghbalzadeh, A. (2018), “Study of inception point, void fraction and pressure over pooled stepped spillways using Flow-3D”, International Journal of Numerical Methods for Heat & Fluid Flow, Vol. 28 No. 4, pp. 982-998. https://doi.org/10.1108/HFF-03-2017-0112

Figure 1- Schematic diagram of pooled stepped spillway conducted by Felder et al. (2012A): Notes: h  step height (10 cm): w pool height (3.1 cm): l horizontal step length (20 cm): lw pool weir length (1.5 cm):  d' is the water depth above the crest; y' is the distance normal to the crest invert
Figure 1- Schematic diagram of pooled stepped spillway conducted by Felder et al. (2012A): Notes: h step height (10 cm): w pool height (3.1 cm): l horizontal step length (20 cm): lw pool weir length (1.5 cm): d’ is the water depth above the crest; y’ is the distance normal to the crest invert
Figure 2- meshing domain and distribution of blocks
Figure 2- meshing domain and distribution of blocks
Figure 3- Comparison of numerical simulation with experimental data by Felder et al. (2012A);  mesh convergence analysis; pooled stepped spillway (slope: 26.6 0 )
Figure 3- Comparison of numerical simulation with experimental data by Felder et al. (2012A); mesh convergence analysis; pooled stepped spillway (slope: 26.6 0 )
Figure 4- Comparison of numerical simulation with experimental data by Felder et al. (2012A);  Flat stepped spillway (slope: 0 26 6. )
Figure 4- Comparison of numerical simulation with experimental data by Felder et al. (2012A); Flat stepped spillway (slope: 0 26 6. )
Figure 5-Comparison of numerical simulation with experimental data by Felder et al. (2012B); pooled  and flat stepped spillways (slope: 0 9.8 )
Figure 5-Comparison of numerical simulation with experimental data by Felder et al. (2012B); pooled and flat stepped spillways (slope: 0 9.8 )
Figure 6- TKE distribution on steps 8, 9 and 10 for four different mesh numbers: 261252 (model 1),  288941 (model 2), 323578 (model 3) and 343154 (model 4)
Figure 6- TKE distribution on steps 8, 9 and 10 for four different mesh numbers: 261252 (model 1), 288941 (model 2), 323578 (model 3) and 343154 (model 4)
Figure 7- Comparison of obtained Void fraction distribution on step 10 in numerical simulation with  experimental work conducted by Felder et al. (2012A); (slope 26.60 )
Figure 7- Comparison of obtained Void fraction distribution on step 10 in numerical simulation with experimental work conducted by Felder et al. (2012A); (slope 26.60 )
Figure 8- Results of inception point of air entrainment in different height of the pools: comparison with  empirical correlations (Eqs 8-9), experimental (Felder et al. (2012A)) and numerical data
Figure 8- Results of inception point of air entrainment in different height of the pools: comparison with empirical correlations (Eqs 8-9), experimental (Felder et al. (2012A)) and numerical data
Figure 9- Void fraction distribution for different pool heights on steps 10; slope 26.6 0
Figure 9- Void fraction distribution for different pool heights on steps 10; slope 26.6 0
Figure 10- Comparison of pressure distribution between numerical simulation and experimental work  conducted by Zhang and Chanson (2016); flat stepped spillway (slope: 0 45 )
Figure 10- Comparison of pressure distribution between numerical simulation and experimental work conducted by Zhang and Chanson (2016); flat stepped spillway (slope: 0 45 )
Figure 11- A comparison of the pressure distribution above the crest of the spillway; B comparison of the  free surface profile along the crest of the spillway.  Note: x' indicates the longitudinal distance from the starting point of the crest.
Figure 11- A comparison of the pressure distribution above the crest of the spillway; B comparison of the free surface profile along the crest of the spillway. Note: x’ indicates the longitudinal distance from the starting point of the crest.
Figure 12- pressure distribution along crest of spillway in different discharges; slope 26.6
Figure 12- pressure distribution along crest of spillway in different discharges; slope 26.6
Figure 13- Pressure distribution near the last step bed for different slopes and discharges: x'' indicatesthe  longitudinal distance from the intersection of the horizontal and vertical faces of step 10; y" is the distance from the intersection of the horizontal and vertical faces in the vertical direction
Figure 13- Pressure distribution near the last step bed for different slopes and discharges: x” indicatesthe longitudinal distance from the intersection of the horizontal and vertical faces of step 10; y” is the distance from the intersection of the horizontal and vertical faces in the vertical direction
Figure 14- Pressure distribution adjacent the vertical face of step 9 for different discharges and slopes
Figure 14- Pressure distribution adjacent the vertical face of step 9 for different discharges and slopes
Table1- Used discharges for assessments of mesh convergence analysis and hydraulic  characteristics
Table1- Used discharges for assessments of mesh convergence analysis and hydraulic characteristics

Conclusion

본 연구에서는 자유표면을 모사하기 위해 VOF 방법과 k -ε (RNG) 난류 모델을 활용하여 FLOW-3D 소프트웨어를 사용하였고, 계단식 배수로의 유동을 모사하기 위한 목적으로 난류 특성을 모사하였다. 얻은 결과는 수치 모델이 시작점 위치, 보이드 비율 및 압력을 적절하게 시뮬레이션했음을 나타냅니다. 풀의 높이는 공기 유입 위치에 미미한 영향을 미치므로 얻은 결과는 이 문서에서 제시된 상관 관계와 잘 일치했습니다. 즉, 사용 가능한 상관 관계를 서로 다른 풀 높이에 사용할 수 있습니다. 공극률의 결과는 스텝 풀 근처의 나프 유동 영역에서 공극율 값이 다른 배출보다 더 큰 것으로 나타났다. 더욱이 고방출량 .0 113m3/s에서 수영장 높이를 변경해도 수영장 표면 근처의 공극률 값에는 영향을 미치지 않았습니다.

낮잠 및 전환 체제의 압력 분포에 대한 0 및 3cm 높이의 수영장 효과는 많은 지점에서 대부분 유사했습니다. 더욱이 조사된 모든 높이에서 여수로의 마루를 따라 부압이 없었습니다. 여수로 끝단의 바닥 부근의 압력 결과는 평평하고 고인 경우 부압이 발생하지 않았음을 나타냅니다. 수직면 부근의 음압은 웅덩이에 비해 평평한 계단형 여수로의 깊이(w=0 cm)의 대부분에서 발생하였다. 또한 더 큰 사면에 대한 풀링 케이스에서 음압이 제거되었습니다. 평단식 여수로에서는 계단의 수직면에 인접한 더 넓은 지역에서 음압이 발생하였기 때문에 이 여수로에서는 고형단식여수로보다 캐비테이션 현상이 발생할 가능성이 더 큽니다.

In this study, the FLOW-3D software was used through utilizing the VOF method and k −ε (RNG) turbulence model in order to simulate free surface, and turbulence characteristics for the purpose of simulating flow over pooled stepped spillway. The results obtained indicated that the numerical model properly simulated the inception point location, void fraction, and pressure. The height of the pools has the insignificant effect on the location of air entrainment, so that obtained results were in good agreement with the correlations presented in this paper. In other words, available correlations can be used for different pool heights. The results of void fraction showed that the void fraction values in nappe flow regime near the step pool were more than the other discharges. Furthermore in high discharge, 0.113m3/s, altering pool height had no effect on the value of void fraction near the pool surface.

The effect of the pools with 0 and 3 cm heights over the pressure distribution in nappe and transition regimes was mostly similar in many points. Furthermore, in all examined heights there was no negative pressure along the crest of the spillway. The pressure results near the bed of the step at the end of the spillway indicated that negative pressure did not occur in the flat and pooled cases. Negative pressure near the vertical face occurred in the most part of the depth in the flat stepped spillway (w=0 cm) in comparison with the pooled case. Also, the negative pressure was eliminated in the pooled case for the larger slopes. Since negative pressure occurred in a larger area adjacent the vertical face of the steps in the flat stepped spillways, it is more likely that cavitation phenomenon occurs in this spillway rather than the pooled stepped spillways.

References

  1. André, S. (2004), “High velocity aerated flows on stepped chutes with macro-roughness elements.” Ph.D. thesis,
    Laboratoire de Constructions Hydraulics (LCH), EPFL, Lausanne, Switzerland, 272 pages.
  2. Attarian, A. Hosseini, Kh. Abdi, H and Hosseini, M. (2014), “The Effect of the Step Height on Energy
    Dissipation in Stepped Spillways Using Numerical Simulation”. Arabian Journal for Science and
    Engineering, 39(4), 2587-2594.
  3. Bombardelli, F.A. Meireles. I. Matos, J. (2011), “Laboratory measurements and multi-block numerical
    simulations of the mean flow and turbulence in the non-aerated skimming flow region of steep stepped
    spillways”. Environmental fluid mechanics, 11(3) 263-288.
  4. Chakib, B. (2013), “Numerical Computation of Inception Point Location for Flat-sloped Stepped Spillway”.
    International Journal of Hydraulic Engineering; 2(3): 47-52.
  5. Chakib, B. Mohammed, H. (2015), “Numerical Simulation of Air Entrainment for Flat-Sloped Stepped Spillway.
    Journal of computational multiphase flows”, Volume 7. Number 1.
  6. Chanson, H. Toombes, L. (2002), “Air–water flows down stepped chutes: turbulence and flow structure
    observations”. International Journal of Multiphase Flow, 28(11) 1737-1761
  7. Chen, Q. Dai, G. Liu, H. (2002), “Volume of Fluid Model for Turbulence Numerical Simulation
    of Stepped Spillway Overflow”. DOI: 10.1061/(ASCE)0733-9429128:7(683).
  8. Cheng, X. Chen, Y. Luo, L. (2006), “Numerical simulation of air-water two-phase flow over stepped spillways”.
    Science in China Series E: Technological Sciences, 49(6), 674-684.
  9. Cheng, X. Luo, L. Zhao, W. (2004), “Study of aeration in the water flow over stepped spillway”. In: Proceedings
    of the world water congress.
  10. Chinnarasri, Ch. Kositgittiwong, D. Julien, Y. (2013), “Model of flow over spillways by computational fluid
    dynamics”. Proceedings of the ICE – Water Management, Volume 167(3) 164 –175.
  11. Dastgheib, A. Niksokhan, M.H. and Nowroozpour, A.R. (2012), “Comparing of Flow Pattern and Energy
    Dissipation over different forms of Stepped Spillway”. World Environmental and Water Resources
    Congress ASCE.
  12. Eghbalzadeh, A. Javan, M. (2012), “Comparison of mixture and VOF models for numerical simulation of air
    entrainment in skimming flow over stepped spillway”. Procedia Engineering, 28. 657-660.
  13. Felder, S, Chanson, H. (2012), “Free-surface Profiles, Velocity and Pressure Distributions on a
    Broad-Crested Weir: a Physical study “Free-surface Profiles, Velocity and Pressure Distributions on a
    Broad-Crested Weir: a Physical study
  14. Felder, S. Fromm, Ch. Chanson, H. (2012B), “Air entrainment and energy dissipation on a 8.9 slope stepped
    spillway with flat and pooled steps”, School of Civil Engineering, The University of Queensland,.
    Brisbane, Australia.
  15. Felder, S. Chanson, H. (2014A), Triple decomposition technique in air–water flows: application to instationary
    flows on a stepped spillway. International Journal of Multiphase Flow, 58, 139-153.
  16. Felder, S. Chanson, H. (2014B), Effects of step pool porosity upon flow aeration and energy dissipation on
    pooled stepped spillways. Journal of Hydraulic Engineering, 140(4), 04014002.
  17. Felder, S. Chanson, H. (2013A), “Air entrainment and energy dissipation on porous pooled stepped spillways”.
    Paper presented at the International Workshop on Hydraulic Design of Low-Head Structures.
  18. Felder, S. Chanson, H. (2013B), “Aeration, flow instabilities, and residual energy on pooled stepped spillways of
    embankment dams”. Journal of irrigation and drainage engineering, 139(10) 880-887.
  19. Felder, S. Guenther, Ph. Chanson, H. (2012A). “Air-water flow properties and energy dissipation on stepped
    spillways: a physical study of several pooled stepped configurations”, School of Civil Engineering, The
    University of Queensland,. Brisbane, Australia.
  20. Flow Science, (2013). “FLOW-3D user’s manual”, version 10.1. Flow Science, Inc, Los Alamos.
  21. Frizell, K.W. Renna, F.M. Matos, J. (2012), “Cavitation potential of flow on stepped spillways”. Journal of
    Hydraulic Engineering, 139(6), 630-636.
  22. Gonzalez, C. (2005), “An experimental study of free-surface aeration on embankment stepped chutes”,
    department of civil engineering, Brisbane, Australia, Phd thesis.
  23. Gonzalez, C.A. Chanson, H. (2008), “Turbulence manipulation in air–water flows on a stepped chute: An
    experimental study”. European Journal of Mechanics-B/Fluids, 27(4), 388-408.
  24. Guenther, Ph.. Felder, S. Chanson, H. (2013), “Flow aeration, cavity processes and energy dissipation on flat and
    pooled stepped spillways for embankments”. Environmental fluid mechanics, 13(5) 503-525.
  25. Hamedi, A. Mansoori, A. Malekmohamadi, I. Roshanaei, H. (2011), “Estimating Energy Dissipation in Stepped
    Spillways with Reverse Inclined Steps and End Sill”. World Environmental and Water Resources
    Congress, ASCE.
  26. Hirt, C.W. (2003), “Modeling Turbulent Entrainment of Air at a Free Surface”. Flow Science Inc.
  27. Hunt, S.L. Kadavy, K.C. (2013), “Inception point for enbankment dam stepped spillway”. J. Hydraul. Eng.,
    139(1), 60–64.
  28. Hunt, S.L. Kadavy, K.C. (2010), “Inception Point Relationship for Flat-Sloped Stepped
    Spillways”. DOI: 10.1061/ASCEHY.1943-7900.0000297.
  29. Matos, J. Quintela, A. (2000), “Air entrainment and safety against cavitation damage in stepped spillways over
    RCC dams. In: Proceeding Intl. Workshop on Hydraulics of Stepped Spillways”, VAW, ETH-Zurich, H.E.
    Minor and W.H. Hager. Balkema. 69–76.
  30. Meireles, I. Matos, J. (2009), “Skimming flow in the nonaerated region of stepped spillways over embankment
    dams”. J. Hydraul. Eng., 135(8), 685–689.
  31. Miang-liang, ZH. Yong-ming, SH. (2008), “Three dimentional simulation of meandering river basin on 3-D
    RNG k − ε turbulence model”. Journal of hydrodynamics, 20(4): 448-455.
  32. Morovati, Kh. Eghbalzadeh, A. Javan, M. (2015), “Numerical investigation of the configuration of the pools on
    the flowPattern passing over pooled stepped spillway in skimming flow regime. Acta Mech, DOI
    10.1007/s00707-015-1444-x
  33. Morovati, Kh. Eghbalzadeh, A. Soori, S. (2016), “Numerical Study of Energy Dissipation of Pooled Stepped
    spillway”. Civil Engineering Journal. Vol. 2, No. 5.
  34. Nikseresht, A.H. Talebbeydokhti, N. and Rezaei, M.J. (2013), “Numerical simulation of two-phase flow on steppool spillways”. Scientia Iranica, A 20 (2), 222–230.
  35. Peyras, L. Royet, P. Degoutte, G. (1990), “Flow and energy dissipation over stepped gabion weirs”. ASCE
    Convention.
  36. Qun, Ch. Guang-qing, D. Feu-qing, Zh. Qing, Y. (2004). “Three-dimensional turbulence numerical simulation of
    a stepped spillway overflow”. Journal of hydrodynamics, Ser. B, 1, 74-79.
  37. Relvas, A. T. Pinheiro, A. N. (2008), Inception point and air concentration in flows on stepped chutes lined with
    wedge-shaped concrete blocks. Journal of Hydraulic Engineering, 134(8), 1042-1051
  38. Sanchez, M. (2000), “Pressure field in skimming flow over a stepped spillways”. In: Proceeding Intl. Workshop
    on Hydraulics of Stepped Spillways, VAW, ETH-Zurich, H.E. Minor and W.H. Hager. Balkema,
    137–146.
  39. Sarfaraz, M. Attari, J. Pfister, M. (2012), “Numerical Computation of Inception Point Location for Steeply
    Sloping Stepped Spillways”. 9th International Congress on Civil Engineering, May 8-10. Isfahan
    University of Technology (IUT), Isfahan, Iran.
  40. Savage, Bruce M. Michael C. Johnson. (2001), “Flow over ogee spillway: Physical and numerical model case
    study.” Journal of Hydraulic Engineering 127.8:640-649.
  41. Shahhedari, H. Jafari Nodoshan, E. Barati, R. Azhdary moghadam, M. (2014). “Discharge coeficient and energy
    dissipation over stepped spillway under skimming flow regime”. KSCE Journal of Civil Engineering, DOI
    10.1007/s12205-013-0749-3.
  42. Tabbara, M. Chatila, J. Awwad, R. (2005), “Computational simulation of flow over stepped spillways”.
    Computers & structures, 83(27) 2215-2224.
  43. Thorwarth, J. (2008), “Hydraulisches Verhalten der Treppengerinne mit eingetieften Stufen—Selbstinduzierte
    Abflussinstationaritäten und Energiedissipation” [Hydraulics of pooled stepped spillways— Self-induced
    unsteady flow and energy dissipation]. Ph.D. thesis, Univ. of Aachen, Aachen, Germany (in German).
  44. WeiLin, XU. ShuJing, LUO, QiuWen, ZH. Jing, LUO. (2015), “Experimental study on pressure and aeration
    characteristics in stepped chute flows. SCIENCE CHINA. Vol.58 No.4: 720–726. doi: 10.1007/s11431-015-
    5783-6.
  45. Xiangju, Ch. Yongcan, C. Lin, L. (2006), “Numerical simulation of air-water two-phase flow over stepped
    spillways”. Science in China Series E: Technological Sciences, 49(6), 674-684.
  46. Zare, K.H. Doering, J.C. (2012), “Inception Point of Air Entrainment and Training Wall
    Characteristics of Baffles and Sills on Stepped Spillways”. DOI: 10.1061/(ASCE)HY
    .1943-7900.0000630.
  47. Zhan, J. Zhang, J. Gong, Y. (2016), “Numerical investigation of air-entrainment in skimming flow over stepped
    spillways”. Theoretical and Applied Mechanics Letters. Volume 6. Pages 139–142.
  48. Zhang, G. Chanson, H. (2016), Hydraulics of the developing flow region of stepped spillways. II: Pressure and
    velocity fields. Journal of Hydraulic Engineering, 142(7).
  49. Zhenwei, M. Zhiyan, Zh. Tao, Zh. (2012), “Numerical Simulation of 3-D Flow Field of Spillway based on VOF
    Method”. Procedia Engineering, 28, 808-812.
  50. Zhi-yong, D. Hun-wei, L.J. (2006), “Numerical simulation of skimming flow over mild stepped channel”.
    Journal of Hydrodynamics, Ser. B, 18(3) 367-371.
  51. ZhongDong, Q. XiaoQing, H. WenXin, H. António, A. (2009), “Numerical simulation and analysis of water
    flow over stepped spillways”. Science in China Series E: Technological Sciences, 52(7) 1958-1965.
Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

AZ91 합금 주물 내 연행 결함에 대한 캐리어 가스의 영향

Effect of carrier gases on the entrainment defects within AZ91 alloy castings

Tian Liab J.M.T.Daviesa Xiangzhen Zhuc
aUniversity of Birmingham, Birmingham B15 2TT, United Kingdom
bGrainger and Worrall Ltd, Bridgnorth WV15 5HP, United Kingdom
cBrunel Centre for Advanced Solidification Technology, Brunel University London, Kingston Ln, London, Uxbridge UB8 3PH, United Kingdom

Abstract

An entrainment defect (also known as a double oxide film defect or bifilm) acts a void containing an entrapped gas when submerged into a light-alloy melt, thus reducing the quality and reproducibility of the final castings. Previous publications, carried out with Al-alloy castings, reported that this trapped gas could be subsequently consumed by the reaction with the surrounding melt, thus reducing the void volume and negative effect of entrainment defects. Compared with Al-alloys, the entrapped gas within Mg-alloy might be more efficiently consumed due to the relatively high reactivity of magnesium. However, research into the entrainment defects within Mg alloys has been significantly limited. In the present work, AZ91 alloy castings were produced under different carrier gas atmospheres (i.e., SF6/CO2, SF6/air). The evolution processes of the entrainment defects contained in AZ91 alloy were suggested according to the microstructure inspections and thermodynamic calculations. The defects formed in the different atmospheres have a similar sandwich-like structure, but their oxide films contained different combinations of compounds. The use of carrier gases, which were associated with different entrained-gas consumption rates, affected the reproducibility of AZ91 castings.

Keywords

Magnesium alloyCastingOxide film, Bifilm, Entrainment defect, Reproducibility

연행 결함(이중 산화막 결함 또는 이중막 결함이라고도 함)은 경합금 용융물에 잠길 때 갇힌 가스를 포함하는 공극으로 작용하여 최종 주물의 품질과 재현성을 저하시킵니다. Al-합금 주조로 수행된 이전 간행물에서는 이 갇힌 가스가 주변 용융물과의 반응에 의해 후속적으로 소모되어 공극 부피와 연행 결함의 부정적인 영향을 줄일 수 있다고 보고했습니다. Al-합금에 비해 마그네슘의 상대적으로 높은 반응성으로 인해 Mg-합금 내에 포집된 가스가 더 효율적으로 소모될 수 있습니다. 그러나 Mg 합금 내 연행 결함에 대한 연구는 상당히 제한적이었습니다. 현재 작업에서 AZ91 합금 주물은 다양한 캐리어 가스 분위기(즉, SF 6 /CO2 , SF 6 / 공기). AZ91 합금에 포함된 엔트레인먼트 결함의 진화 과정은 미세조직 검사 및 열역학적 계산에 따라 제안되었습니다. 서로 다른 분위기에서 형성된 결함은 유사한 샌드위치 구조를 갖지만 산화막에는 서로 다른 화합물 조합이 포함되어 있습니다. 다른 동반 가스 소비율과 관련된 운반 가스의 사용은 AZ91 주물의 재현성에 영향을 미쳤습니다.

키워드

마그네슘 합금주조Oxide film, Bifilm, Entrainment 불량, 재현성

1 . 소개

지구상에서 가장 가벼운 구조용 금속인 마그네슘은 지난 수십 년 동안 가장 매력적인 경금속 중 하나가 되었습니다. 결과적으로 마그네슘 산업은 지난 20년 동안 급속한 발전을 경험했으며 [1 , 2] , 이는 전 세계적으로 Mg 합금에 대한 수요가 크게 증가했음을 나타냅니다. 오늘날 Mg 합금의 사용은 자동차, 항공 우주, 전자 등의 분야에서 볼 수 있습니다. [3 , 4] . Mg 금속의 전 세계 소비는 특히 자동차 산업에서 앞으로 더욱 증가할 것으로 예측되었습니다. 기존 자동차와 전기 자동차 모두의 에너지 효율성 요구 사항이 설계를 경량화하도록 더욱 밀어붙이기 때문입니다 [3 , 56] .

Mg 합금에 대한 수요의 지속적인 성장은 Mg 합금 주조의 품질 및 기계적 특성 개선에 대한 광범위한 관심을 불러일으켰습니다. Mg 합금 주조 공정 동안 용융물의 표면 난류는 소량의 주변 대기를 포함하는 이중 표면 필름의 포획으로 이어질 수 있으므로 동반 결함(이중 산화막 결함 또는 이중막 결함이라고도 함)을 형성합니다. ) [7] , [8] , [9] , [10] . 무작위 크기, 수량, 방향 및 연행 결함의 배치는 주조 특성의 변화와 관련된 중요한 요인으로 널리 받아들여지고 있습니다 [7] . 또한 Peng et al. [11]AZ91 합금 용융물에 동반된 산화물 필름이 Al 8 Mn 5 입자에 대한 필터 역할을 하여 침전될 때 가두는 것을 발견했습니다 . Mackie et al. [12]는 또한 동반된 산화막이 금속간 입자를 트롤(trawl)하는 작용을 하여 입자가 클러스터링되어 매우 큰 결함을 형성할 수 있다고 제안했습니다. 금속간 화합물의 클러스터링은 비말동반 결함을 주조 특성에 더 해롭게 만들었습니다.

연행 결함에 관한 이전 연구의 대부분은 Al-합금에 대해 수행되었으며 [7 , [13] , [14] , [15] , [16] , [17] , [18] 몇 가지 잠재적인 방법이 제안되었습니다. 알루미늄 합금 주물의 품질에 대한 부정적인 영향을 줄이기 위해. Nyahumwa et al., [16] 은 연행 결함 내의 공극 체적이 열간 등방압 압축(HIP) 공정에 의해 감소될 수 있음을 보여줍니다. Campbell [7] 은 결함 내부의 동반된 가스가 주변 용융물과의 반응으로 인해 소모될 수 있다고 제안했으며, 이는 Raiszedeh와 Griffiths [19]에 의해 추가로 확인되었습니다 ..혼입 가스 소비가 Al-합금 주물의 기계적 특성에 미치는 영향은 [8 , 9]에 의해 조사되었으며 , 이는 혼입 가스의 소비가 주조 재현성의 개선을 촉진함을 시사합니다.

Al-합금 내 결함에 대한 조사와 비교하여 Mg-합금 내 연행 결함에 대한 연구는 상당히 제한적입니다. 연행 결함의 존재는 Mg 합금 주물 [20 , 21] 에서 입증 되었지만 그 거동, 진화 및 연행 가스 소비는 여전히 명확하지 않습니다.

Mg 합금 주조 공정에서 용융물은 일반적으로 마그네슘 점화를 피하기 위해 커버 가스로 보호됩니다. 따라서 모래 또는 매몰 몰드의 공동은 용융물을 붓기 전에 커버 가스로 세척해야 합니다 [22] . 따라서, Mg 합금 주물 내의 연행 가스는 공기만이 아니라 주조 공정에 사용되는 커버 가스를 포함해야 하며, 이는 구조 및 해당 연행 결함의 전개를 복잡하게 만들 수 있습니다.

SF 6 은 Mg 합금 주조 공정에 널리 사용되는 대표적인 커버 가스입니다 [23] , [24] , [25] . 이 커버 가스는 유럽의 마그네슘 합금 주조 공장에서 사용하도록 제한되었지만 상업 보고서에 따르면 이 커버는 전 세계 마그네슘 합금 산업, 특히 다음과 같은 글로벌 마그네슘 합금 생산을 지배한 국가에서 여전히 인기가 있습니다. 중국, 브라질, 인도 등 [26] . 또한, 최근 학술지 조사에서도 이 커버가스가 최근 마그네슘 합금 연구에서 널리 사용된 것으로 나타났다 [27] . SF 6 커버 가스 의 보호 메커니즘 (즉, 액체 Mg 합금과 SF 6 사이의 반응Cover gas)에 대한 연구는 여러 선행연구자들에 의해 이루어졌으나 표면 산화막의 형성과정이 아직 명확하게 밝혀지지 않았으며, 일부 발표된 결과들도 상충되고 있다. 1970년대 초 Fruehling [28] 은 SF 6 아래에 형성된 표면 피막이 주로 미량의 불화물과 함께 MgO 임을 발견 하고 SF 6 이 Mg 합금 표면 피막에 흡수 된다고 제안했습니다 . Couling [29] 은 흡수된 SF 6 이 Mg 합금 용융물과 반응하여 MgF 2 를 형성함을 추가로 확인했습니다 . 지난 20년 동안 아래에 자세히 설명된 것처럼 Mg 합금 표면 필름의 다양한 구조가 보고되었습니다.(1)

단층 필름 . Cashion [30 , 31] 은 X선 광전자 분광법(XPS)과 오제 분광법(AES)을 사용하여 표면 필름을 MgO 및 MgF 2 로 식별했습니다 . 그는 또한 필름의 구성이 두께와 전체 실험 유지 시간에 걸쳐 일정하다는 것을 발견했습니다. Cashion이 관찰한 필름은 10분에서 100분의 유지 시간으로 생성된 단층 구조를 가졌다.(2)

이중층 필름 . Aarstad et. al [32] 은 2003년에 이중층 표면 산화막을 보고했습니다. 그들은 예비 MgO 막에 부착된 잘 분포된 여러 MgF 2 입자를 관찰 하고 전체 표면적의 25-50%를 덮을 때까지 성장했습니다. 외부 MgO 필름을 통한 F의 내부 확산은 진화 과정의 원동력이었습니다. 이 이중층 구조는 Xiong의 그룹 [25 , 33] 과 Shih et al. 도 지지했습니다 . [34] .(삼)

트리플 레이어 필름 . 3층 필름과 그 진화 과정은 Pettersen [35]에 의해 2002년에 보고되었습니다 . Pettersen은 초기 표면 필름이 MgO 상이었고 F의 내부 확산에 의해 점차적으로 안정적인 MgF 2 상 으로 진화한다는 것을 발견했습니다 . 두꺼운 상부 및 하부 MgF 2 층.(4)

산화물 필름은 개별 입자로 구성 됩니다. Wang et al [36] 은 Mg-alloy 표면 필름을 SF 6 커버 가스 하에서 용융물에 교반 한 다음 응고 후 동반된 표면 필름을 검사했습니다. 그들은 동반된 표면 필름이 다른 연구자들이 보고한 보호 표면 필름처럼 계속되지 않고 개별 입자로 구성된다는 것을 발견했습니다. 젊은 산화막은 MgO 나노 크기의 산화물 입자로 구성되어 있는 반면, 오래된 산화막은 한쪽 면에 불화물과 질화물이 포함된 거친 입자(평균 크기 약 1μm)로 구성되어 있습니다.

Mg 합금 용융 표면의 산화막 또는 동반 가스는 모두 액체 Mg 합금과 커버 가스 사이의 반응으로 인해 형성되므로 Mg 합금 표면막에 대한 위에서 언급한 연구는 진화에 대한 귀중한 통찰력을 제공합니다. 연행 결함. 따라서 SF 6 커버 가스 의 보호 메커니즘 (즉, Mg-합금 표면 필름의 형성)은 해당 동반 결함의 잠재적인 복잡한 진화 과정을 나타냅니다.

그러나 Mg 합금 용융물에 표면 필름을 형성하는 것은 용융물에 잠긴 동반된 가스의 소비와 다른 상황에 있다는 점에 유의해야 합니다. 예를 들어, 앞서 언급한 연구에서 표면 성막 동안 충분한 양의 커버 가스가 담지되어 커버 가스의 고갈을 억제했습니다. 대조적으로, Mg 합금 용융물 내의 동반된 가스의 양은 유한하며, 동반된 가스는 완전히 고갈될 수 있습니다. Mirak [37] 은 3.5% SF 6 /기포를 특별히 설계된 영구 금형에서 응고되는 순수한 Mg 합금 용융물에 도입했습니다. 기포가 완전히 소모되었으며, 해당 산화막은 MgO와 MgF 2 의 혼합물임을 알 수 있었다.. 그러나 Aarstad [32] 및 Xiong [25 , 33]에 의해 관찰된 MgF 2 스팟 과 같은 핵 생성 사이트 는 관찰되지 않았습니다. Mirak은 또한 조성 분석을 기반으로 산화막에서 MgO 이전에 MgF 2 가 형성 되었다고 추측했는데 , 이는 이전 문헌에서 보고된 표면 필름 형성 과정(즉, MgF 2 이전에 형성된 MgO)과 반대 입니다. Mirak의 연구는 동반된 가스의 산화막 형성이 표면막의 산화막 형성과 상당히 다를 수 있음을 나타내었지만 산화막의 구조와 진화에 대해서는 밝히지 않았습니다.

또한 커버 가스에 캐리어 가스를 사용하는 것도 커버 가스와 액체 Mg 합금 사이의 반응에 영향을 미쳤습니다. SF 6 /air 는 용융 마그네슘의 점화를 피하기 위해 SF 6 /CO 2 운반 가스 [38] 보다 더 높은 함량의 SF 6을 필요로 하여 다른 가스 소비율을 나타냅니다. Liang et.al [39] 은 CO 2 가 캐리어 가스로 사용될 때 표면 필름에 탄소가 형성된다고 제안했는데 , 이는 SF 6 /air 에서 형성된 필름과 다릅니다 . Mg 연소 [40]에 대한 조사 에서 Mg 2 C 3 검출이 보고되었습니다.CO 2 연소 후 Mg 합금 샘플 에서 이는 Liang의 결과를 뒷받침할 뿐만 아니라 이중 산화막 결함에서 Mg 탄화물의 잠재적 형성을 나타냅니다.

여기에 보고된 작업은 다양한 커버 가스(즉, SF 6 /air 및 SF 6 /CO 2 )로 보호되는 AZ91 Mg 합금 주물에서 형성된 연행 결함의 거동과 진화에 대한 조사 입니다. 이러한 캐리어 가스는 액체 Mg 합금에 대해 다른 보호성을 가지며, 따라서 상응하는 동반 가스의 다른 소비율 및 발생 프로세스와 관련될 수 있습니다. AZ91 주물의 재현성에 대한 동반 가스 소비의 영향도 연구되었습니다.

2 . 실험

2.1 . 용융 및 주조

3kg의 AZ91 합금을 700 ± 5 °C의 연강 도가니에서 녹였습니다. AZ91 합금의 조성은 표 1 에 나타내었다 . 가열하기 전에 잉곳 표면의 모든 산화물 스케일을 기계가공으로 제거했습니다. 사용 된 커버 가스는 0.5 %이었다 SF 6 / 공기 또는 0.5 % SF 6 / CO 2 (부피. %) 다른 주물 6L / 분의 유량. 용융물은 15분 동안 0.3L/min의 유속으로 아르곤으로 가스를 제거한 다음 [41 , 42] , 모래 주형에 부었습니다. 붓기 전에 샌드 몰드 캐비티를 20분 동안 커버 가스로 플러싱했습니다 [22] . 잔류 용융물(약 1kg)이 도가니에서 응고되었습니다.

표 1 . 본 연구에 사용된 AZ91 합금의 조성(wt%).

아연미네소타마그네슘
9.40.610.150.020.0050.0017잔여

그림 1 (a)는 러너가 있는 주물의 치수를 보여줍니다. 탑 필링 시스템은 최종 주물에서 연행 결함을 생성하기 위해 의도적으로 사용되었습니다. Green과 Campbell [7 , 43] 은 탑 필링 시스템이 바텀 필링 시스템에 비해 주조 과정에서 더 많은 연행 현상(즉, 이중 필름)을 유발한다고 제안했습니다. 이 금형의 용융 흐름 시뮬레이션(Flow-3D 소프트웨어)은 연행 현상에 관한 Reilly의 모델 [44] 을 사용하여 최종 주조에 많은 양의 이중막이 포함될 것이라고 예측했습니다( 그림 1 에서 검은색 입자로 표시됨) . NS).

그림 1

수축 결함은 또한 주물의 기계적 특성과 재현성에 영향을 미칩니다. 이 연구는 주조 품질에 대한 이중 필름의 영향에 초점을 맞추었기 때문에 수축 결함이 발생하지 않도록 금형을 의도적으로 설계했습니다. ProCAST 소프트웨어를 사용한 응고 시뮬레이션은 그림 1c 와 같이 최종 주조에 수축 결함이 포함되지 않음을 보여주었습니다 . 캐스팅 건전함도 테스트바 가공 전 실시간 X-ray를 통해 확인했다.

모래 주형은 1wt를 함유한 수지 결합된 규사로 만들어졌습니다. % PEPSET 5230 수지 및 1wt. % PEPSET 5112 촉매. 모래는 또한 억제제로 작용하기 위해 2중량%의 Na 2 SiF 6 을 함유했습니다 .. 주입 온도는 700 ± 5 °C였습니다. 응고 후 러너바의 단면을 Sci-Lab Analytical Ltd로 보내 H 함량 분석(LECO 분석)을 하였고, 모든 H 함량 측정은 주조 공정 후 5일째에 실시하였다. 각각의 주물은 인장 강도 시험을 위해 클립 신장계가 있는 Zwick 1484 인장 시험기를 사용하여 40개의 시험 막대로 가공되었습니다. 파손된 시험봉의 파단면을 주사전자현미경(SEM, Philips JEOL7000)을 이용하여 가속전압 5~15kV로 조사하였다. 파손된 시험 막대, 도가니에서 응고된 잔류 Mg 합금 및 주조 러너를 동일한 SEM을 사용하여 단면화하고 연마하고 검사했습니다. CFEI Quanta 3D FEG FIB-SEM을 사용하여 FIB(집속 이온 빔 밀링 기술)에 의해 테스트 막대 파괴 표면에서 발견된 산화막의 단면을 노출했습니다. 분석에 필요한 산화막은 백금층으로 코팅하였다. 그런 다음 30kV로 가속된 갈륨 이온 빔이 산화막의 단면을 노출시키기 위해 백금 코팅 영역을 둘러싼 재료 기판을 밀링했습니다. 산화막 단면의 EDS 분석은 30kV의 가속 전압에서 FIB 장비를 사용하여 수행되었습니다.

2.2 . 산화 세포

전술 한 바와 같이, 몇몇 최근 연구자들은 마그네슘 합금의 용탕 표면에 형성된 보호막 조사 [38 , 39 , [46] , [47] , [48] , [49] , [50] , [51] , [52 ] . 이 실험 동안 사용된 커버 가스의 양이 충분하여 커버 가스에서 불화물의 고갈을 억제했습니다. 이 섹션에서 설명하는 실험은 엔트레인먼트 결함의 산화막의 진화를 연구하기 위해 커버 가스의 공급을 제한하는 밀봉된 산화 셀을 사용했습니다. 산화 셀에 포함된 커버 가스는 큰 크기의 “동반된 기포”로 간주되었습니다.

도 2에 도시된 바와 같이 , 산화셀의 본체는 내부 길이가 400mm, 내경이 32mm인 폐쇄형 연강관이었다. 수냉식 동관을 전지의 상부에 감았습니다. 튜브가 가열될 때 냉각 시스템은 상부와 하부 사이에 온도 차이를 만들어 내부 가스가 튜브 내에서 대류하도록 했습니다. 온도는 도가니 상단에 위치한 K형 열전대로 모니터링했습니다. Nieet al. [53] 은 Mg 합금 용융물의 표면 피막을 조사할 때 SF 6 커버 가스가 유지로의 강철 벽과 반응할 것이라고 제안했습니다 . 이 반응을 피하기 위해 강철 산화 전지의 내부 표면(그림 2 참조)) 및 열전대의 상반부는 질화붕소로 코팅되었습니다(Mg 합금은 질화붕소와 ​​접촉하지 않았습니다).

그림 2

실험 중에 고체 AZ91 합금 블록을 산화 셀 바닥에 위치한 마그네시아 도가니에 넣었습니다. 전지는 1L/min의 가스 유속으로 전기 저항로에서 100℃로 가열되었다. 원래의 갇힌 대기(즉, 공기)를 대체하기 위해 셀을 이 온도에서 20분 동안 유지했습니다. 그런 다음, 산화 셀을 700°C로 더 가열하여 AZ91 샘플을 녹였습니다. 그런 다음 가스 입구 및 출구 밸브가 닫혀 제한된 커버 가스 공급 하에서 산화를 위한 밀폐된 환경이 생성되었습니다. 그런 다음 산화 전지를 5분 간격으로 5분에서 30분 동안 700 ± 10°C에서 유지했습니다. 각 유지 시간이 끝날 때 세포를 물로 켄칭했습니다. 실온으로 냉각한 후 산화된 샘플을 절단하고 연마한 다음 SEM으로 검사했습니다.

3 . 결과

3.1 . SF 6 /air 에서 형성된 엔트레인먼트 결함의 구조 및 구성

0.5 % SF의 커버 가스 하에서 AZ91 주물에 형성된 유입 결함의 구조 및 조성 6 / 공기는 SEM 및 EDS에 의해 관찰되었다. 결과는 그림 3에 스케치된 엔트레인먼트 결함의 두 가지 유형이 있음을 나타냅니다 . (1) 산화막이 전통적인 단층 구조를 갖는 유형 A 결함 및 (2) 산화막이 2개 층을 갖는 유형 B 결함. 이러한 결함의 세부 사항은 다음에 소개되었습니다. 여기에서 비말동반 결함은 생물막 또는 이중 산화막으로도 알려져 있기 때문에 B형 결함의 산화막은 본 연구에서 “다층 산화막” 또는 “다층 구조”로 언급되었습니다. “이중 산화막 결함의 이중층 산화막”과 같은 혼란스러운 설명을 피하기 위해.

그림 3

그림 4 (ab)는 약 0.4μm 두께의 조밀한 단일층 산화막을 갖는 Type A 결함을 보여줍니다. 이 필름에서 산소, 불소, 마그네슘 및 알루미늄이 검출되었습니다( 그림 4c). 산화막은 마그네슘과 알루미늄의 산화물과 불화물의 혼합물로 추측됩니다. 불소의 검출은 동반된 커버 가스가 이 결함의 형성에 포함되어 있음을 보여주었습니다. 즉, Fig. 4 (a)에 나타난 기공 은 수축결함이나 수소기공도가 아니라 연행결함이었다. 알루미늄의 검출은 Xiong과 Wang의 이전 연구 [47 , 48] 와 다르며 , SF 6으로 보호된 AZ91 용융물의 표면 필름에 알루미늄이 포함되어 있지 않음을 보여주었습니다.커버 가스. 유황은 원소 맵에서 명확하게 인식할 수 없었지만 해당 ESD 스펙트럼에서 S-피크가 있었습니다.

그림 4

도 5 (ab)는 다층 산화막을 갖는 Type B 엔트레인먼트 결함을 나타낸다. 산화막의 조밀한 외부 층은 불소와 산소가 풍부하지만( 그림 5c) 상대적으로 다공성인 내부 층은 산소만 풍부하고(즉, 불소가 부족) 부분적으로 함께 성장하여 샌드위치 모양을 형성합니다. 구조. 따라서 외층은 불화물과 산화물의 혼합물이며 내층은 주로 산화물로 추정된다. 황은 EDX 스펙트럼에서만 인식될 수 있었고 요소 맵에서 명확하게 식별할 수 없었습니다. 이는 커버 가스의 작은 S 함량(즉, SF 6 의 0.5% 부피 함량 때문일 수 있음)커버 가스). 이 산화막에서는 이 산화막의 외층에 알루미늄이 포함되어 있지만 내층에서는 명확하게 검출할 수 없었다. 또한 Al의 분포가 고르지 않은 것으로 보입니다. 결함의 우측에는 필름에 알루미늄이 존재하지만 그 농도는 매트릭스보다 높은 것으로 식별할 수 없음을 알 수 있다. 그러나 결함의 왼쪽에는 알루미늄 농도가 훨씬 높은 작은 영역이 있습니다. 이러한 알루미늄의 불균일한 분포는 다른 결함(아래 참조)에서도 관찰되었으며, 이는 필름 내부 또는 아래에 일부 산화물 입자가 형성된 결과입니다.

그림 5

무화과 도 4 및 5 는 SF 6 /air 의 커버 가스 하에 주조된 AZ91 합금 샘플에서 형성된 연행 결함의 횡단면 관찰을 나타낸다 . 2차원 단면에서 관찰된 수치만으로 연행 결함을 특성화하는 것만으로는 충분하지 않습니다. 더 많은 이해를 돕기 위해 테스트 바의 파단면을 관찰하여 엔트레인먼트 결함(즉, 산화막)의 표면을 더 연구했습니다.

Fig. 6 (a)는 SF 6 /air 에서 생산된 AZ91 합금 인장시험봉의 파단면을 보여준다 . 파단면의 양쪽에서 대칭적인 어두운 영역을 볼 수 있습니다. 그림 6 (b)는 어두운 영역과 밝은 영역 사이의 경계를 보여줍니다. 밝은 영역은 들쭉날쭉하고 부서진 특징으로 구성되어 있는 반면, 어두운 영역의 표면은 비교적 매끄럽고 평평했습니다. 또한 EDS 결과( Fig. 6 c-d 및 Table 2) 불소, 산소, 황 및 질소는 어두운 영역에서만 검출되었으며, 이는 어두운 영역이 용융물에 동반된 표면 보호 필름임을 나타냅니다. 따라서 어두운 영역은 대칭적인 특성을 고려할 때 연행 결함이라고 제안할 수 있습니다. Al-합금 주조물의 파단면에서 유사한 결함이 이전에 보고되었습니다 [7] . 질화물은 테스트 바 파단면의 산화막에서만 발견되었지만 그림 1과 그림 4에 표시된 단면 샘플에서는 검출되지 않았습니다 4 및 5 . 근본적인 이유는 이러한 샘플에 포함된 질화물이 샘플 연마 과정에서 가수분해되었을 수 있기 때문입니다 [54] .

그림 6

표 2 . EDS 결과(wt.%)는 그림 6에 표시된 영역에 해당합니다 (커버 가스: SF 6 /공기).

영형마그네슘NS아연NSNS
그림 6 (b)의 어두운 영역3.481.3279.130.4713.630.570.080.73
그림 6 (b)의 밝은 영역3.5884.4811.250.68

도 1 및 도 2에 도시된 결함의 단면 관찰과 함께 도 4 및 도 5 를 참조하면, 인장 시험봉에 포함된 연행 결함의 구조를 도 6 (e) 와 같이 스케치하였다 . 결함에는 산화막으로 둘러싸인 동반된 가스가 포함되어 있어 테스트 바 내부에 보이드 섹션이 생성되었습니다. 파괴 과정에서 결함에 인장력이 가해지면 균열이 가장 약한 경로를 따라 전파되기 때문에 보이드 섹션에서 균열이 시작되어 연행 결함을 따라 전파됩니다 [55] . 따라서 최종적으로 시험봉이 파단되었을 때 Fig. 6 (a) 와 같이 시험봉의 양 파단면에 연행결함의 산화피막이 나타났다 .

3.2 . SF 6 /CO 2 에 형성된 연행 결함의 구조 및 조성

SF 6 /air 에서 형성된 엔트레인먼트 결함과 유사하게, 0.5% SF 6 /CO 2 의 커버 가스 아래에서 형성된 결함 도 두 가지 유형의 산화막(즉, 단층 및 다층 유형)을 가졌다. 도 7 (a)는 다층 산화막을 포함하는 엔트레인먼트 결함의 예를 도시한다. 결함에 대한 확대 관찰( 그림 7b )은 산화막의 내부 층이 함께 성장하여 SF 6 /air 의 분위기에서 형성된 결함과 유사한 샌드위치 같은 구조를 나타냄을 보여줍니다 ( 그림 7b). 5 나 ). EDS 스펙트럼( 그림 7c) 이 샌드위치형 구조의 접합부(내층)는 주로 산화마그네슘을 함유하고 있음을 보여주었다. 이 EDS 스펙트럼에서는 불소, 황, 알루미늄의 피크가 확인되었으나 그 양은 상대적으로 적었다. 대조적으로, 산화막의 외부 층은 조밀하고 불화물과 산화물의 혼합물로 구성되어 있습니다( 그림 7d-e).

그림 7

Fig. 8 (a)는 0.5%SF 6 /CO 2 분위기에서 제작된 AZ91 합금 인장시험봉의 파단면의 연행결함을 보여준다 . 상응하는 EDS 결과(표 3)는 산화막이 불화물과 산화물을 함유함을 보여주었다. 황과 질소는 검출되지 않았습니다. 게다가, 확대 관찰(  8b)은 산화막 표면에 반점을 나타내었다. 반점의 직경은 수백 나노미터에서 수 마이크론 미터까지 다양했습니다.

그림 8

산화막의 구조와 조성을 보다 명확하게 나타내기 위해 테스트 바 파단면의 산화막 단면을 FIB 기법을 사용하여 현장에서 노출시켰다( 그림 9 ). 도 9a에 도시된 바와 같이 , 백금 코팅층과 Mg-Al 합금 기재 사이에 연속적인 산화피막이 발견되었다. 그림 9 (bc)는 다층 구조( 그림 9c 에서 빨간색 상자로 표시)를 나타내는 산화막에 대한 확대 관찰을 보여줍니다 . 바닥층은 불소와 산소가 풍부하고 불소와 산화물의 혼합물이어야 합니다 . 5 와 7, 유일한 산소가 풍부한 최상층은 도 1 및 도 2에 도시 된 “내층”과 유사하였다 5 및 7 .

그림 9

연속 필름을 제외하고 도 9 에 도시된 바와 같이 연속 필름 내부 또는 하부에서도 일부 개별 입자가 관찰되었다 . 그림 9( b) 의 산화막 좌측에서 Al이 풍부한 입자가 검출되었으며, 마그네슘과 산소 원소도 풍부하게 함유하고 있어 스피넬 Mg 2 AlO 4 로 추측할 수 있다 . 이러한 Mg 2 AlO 4 입자의 존재는 Fig. 5 와 같이 관찰된 필름의 작은 영역에 높은 알루미늄 농도와 알루미늄의 불균일한 분포의 원인이 된다 .(씨). 여기서 강조되어야 할 것은 연속 산화막의 바닥층의 다른 부분이 이 Al이 풍부한 입자보다 적은 양의 알루미늄을 함유하고 있지만, 그림 9c는 이 바닥층의 알루미늄 양이 여전히 무시할 수 없는 수준임을 나타냅니다 . , 특히 필름의 외층과 비교할 때. 도 9b에 도시된 산화막의 우측 아래에서 입자가 검출되어 Mg와 O가 풍부하여 MgO인 것으로 추측되었다. Wang의 결과에 따르면 [56], Mg 용융물과 Mg 증기의 산화에 의해 Mg 용융물의 표면에 많은 이산 MgO 입자가 형성될 수 있다. 우리의 현재 연구에서 관찰된 MgO 입자는 같은 이유로 인해 형성될 수 있습니다. 실험 조건의 차이로 인해 더 적은 Mg 용융물이 기화되거나 O2와 반응할 수 있으므로 우리 작업에서 형성되는 MgO 입자는 소수에 불과합니다. 또한 필름에서 풍부한 탄소가 발견되어 CO 2 가 용융물과 반응하여 탄소 또는 탄화물을 형성할 수 있음을 보여줍니다 . 이 탄소 농도는 표 3에 나타낸 산화막의 상대적으로 높은 탄소 함량 (즉, 어두운 영역) 과 일치하였다 . 산화막 옆 영역.

표 3 . 도 8에 도시된 영역에 상응하는 EDS 결과(wt.%) (커버 가스: SF 6 / CO 2 ).

영형마그네슘NS아연NSNS
그림 8 (a)의 어두운 영역7.253.6469.823.827.030.86
그림 8 (a)의 밝은 영역2.100.4482.8313.261.36

테스트 바 파단면( 도 9 ) 에서 산화막의 이 단면 관찰은 도 6 (e)에 도시된 엔트레인먼트 결함의 개략도를 추가로 확인했다 . SF 6 /CO 2 와 SF 6 /air 의 서로 다른 분위기에서 형성된 엔트레인먼트 결함 은 유사한 구조를 가졌지만 그 조성은 달랐다.

3.3 . 산화 전지에서 산화막의 진화

섹션 3.1 및 3.2 의 결과 는 SF 6 /air 및 SF 6 /CO 2 의 커버 가스 아래에서 AZ91 주조에서 형성된 연행 결함의 구조 및 구성을 보여줍니다 . 산화 반응의 다른 단계는 연행 결함의 다른 구조와 조성으로 이어질 수 있습니다. Campbell은 동반된 가스가 주변 용융물과 반응할 수 있다고 추측했지만 Mg 합금 용융물과 포획된 커버 가스 사이에 반응이 발생했다는 보고는 거의 없습니다. 이전 연구자들은 일반적으로 개방된 환경에서 Mg 합금 용융물과 커버 가스 사이의 반응에 초점을 맞췄습니다 [38 , 39 , [46] , [47][48] , [49] , [50] , [51] , [52] , 이는 용융물에 갇힌 커버 가스의 상황과 다릅니다. AZ91 합금에서 엔트레인먼트 결함의 형성을 더 이해하기 위해 엔트레인먼트 결함의 산화막의 진화 과정을 산화 셀을 사용하여 추가로 연구했습니다.

.도 10 (a 및 d) 0.5 % 방송 SF 보호 산화 셀에서 5 분 동안 유지 된 표면 막 (6) / 공기. 불화물과 산화물(MgF 2 와 MgO) 로 이루어진 단 하나의 층이 있었습니다 . 이 표면 필름에서. 황은 EDS 스펙트럼에서 검출되었지만 그 양이 너무 적어 원소 맵에서 인식되지 않았습니다. 이 산화막의 구조 및 조성은 도 4 에 나타낸 엔트레인먼트 결함의 단층막과 유사하였다 .

그림 10

10분의 유지 시간 후, 얇은 (O,S)가 풍부한 상부층(약 700nm)이 예비 F-농축 필름에 나타나 그림 10 (b 및 e) 에서와 같이 다층 구조를 형성했습니다 . ). (O, S)가 풍부한 최상층의 두께는 유지 시간이 증가함에 따라 증가했습니다. Fig. 10 (c, f) 에서 보는 바와 같이 30분간 유지한 산화막도 다층구조를 가지고 있으나 (O,S)가 풍부한 최상층(약 2.5μm)의 두께가 10분 산화막의 그것. 도 10 (bc) 에 도시 된 다층 산화막 은 도 5에 도시된 샌드위치형 결함의 막과 유사한 외관을 나타냈다 .

도 10에 도시된 산화막의 상이한 구조는 커버 가스의 불화물이 AZ91 합금 용융물과의 반응으로 인해 우선적으로 소모될 것임을 나타내었다. 불화물이 고갈된 후, 잔류 커버 가스는 액체 AZ91 합금과 추가로 반응하여 산화막에 상부 (O, S)가 풍부한 층을 형성했습니다. 따라서 도 1 및 도 3에 도시된 연행 결함의 상이한 구조 및 조성 4 와 5 는 용융물과 갇힌 커버 가스 사이의 진행 중인 산화 반응 때문일 수 있습니다.

이 다층 구조는 Mg 합금 용융물에 형성된 보호 표면 필름에 관한 이전 간행물 [38 , [46] , [47] , [48] , [49] , [50] , [51] 에서 보고되지 않았습니다 . . 이는 이전 연구원들이 무제한의 커버 가스로 실험을 수행했기 때문에 커버 가스의 불화물이 고갈되지 않는 상황을 만들었기 때문일 수 있습니다. 따라서 엔트레인먼트 결함의 산화피막은 도 10에 도시된 산화피막과 유사한 거동특성을 가지나 [38 ,[46] , [47] , [48] , [49] , [50] , [51] .

SF 유지 산화막와 마찬가지로 6 / 공기, SF에 형성된 산화물 막 (6) / CO 2는 또한 세포 산화 다른 유지 시간과 다른 구조를 가지고 있었다. .도 11 (a)는 AZ91 개최 산화막, 0.5 %의 커버 가스 하에서 SF 표면 용융 도시 6 / CO 2, 5 분. 이 필름은 MgF 2 로 이루어진 단층 구조를 가졌다 . 이 영화에서는 MgO의 존재를 확인할 수 없었다. 30분의 유지 시간 후, 필름은 다층 구조를 가졌다; 내부 층은 조밀하고 균일한 외관을 가지며 MgF 2 로 구성 되고 외부 층은 MgF 2 혼합물및 MgO. 0.5%SF 6 /air 에서 형성된 표면막과 다른 이 막에서는 황이 검출되지 않았다 . 따라서, 0.5%SF 6 /CO 2 의 커버 가스 내의 불화물 도 막 성장 과정의 초기 단계에서 우선적으로 소모되었다. SF 6 /air 에서 형성된 막과 비교하여 SF 6 /CO 2 에서 형성된 막에서 MgO 는 나중에 나타났고 황화물은 30분 이내에 나타나지 않았다. 이는 SF 6 /air 에서 필름의 형성과 진화 가 SF 6 /CO 2 보다 빠르다 는 것을 의미할 수 있습니다 . CO 2 후속적으로 용융물과 반응하여 MgO를 형성하는 반면, 황 함유 화합물은 커버 가스에 축적되어 반응하여 매우 늦은 단계에서 황화물을 형성할 수 있습니다(산화 셀에서 30분 후).

그림 11

4 . 논의

4.1 . SF 6 /air 에서 형성된 연행 결함의 진화

Outokumpu HSC Chemistry for Windows( http://www.hsc-chemistry.net/ )의 HSC 소프트웨어를 사용하여 갇힌 기체와 액체 AZ91 합금 사이에서 발생할 수 있는 반응을 탐색하는 데 필요한 열역학 계산을 수행했습니다. 계산에 대한 솔루션은 소량의 커버 가스(즉, 갇힌 기포 내의 양)와 AZ91 합금 용융물 사이의 반응 과정에서 어떤 생성물이 가장 형성될 가능성이 있는지 제안합니다.

실험에서 압력은 1기압으로, 온도는 700°C로 설정했습니다. 커버 가스의 사용량은 7 × 10으로 가정 하였다 -7  약 0.57 cm의 양으로 kg 3 (3.14 × 10 -6  0.5 % SF위한 kmol) 6 / 공기, 0.35 cm (3) (3.12 × 10 – 8  kmol) 0.5%SF 6 /CO 2 . 포획된 가스와 접촉하는 AZ91 합금 용융물의 양은 모든 반응을 완료하기에 충분한 것으로 가정되었습니다. SF 6 의 분해 생성물 은 SF 5 , SF 4 , SF 3 , SF 2 , F 2 , S(g), S 2(g) 및 F(g) [57] , [58] , [59] , [60] .

그림 12 는 AZ91 합금과 0.5%SF 6 /air 사이의 반응에 대한 열역학적 계산의 평형 다이어그램을 보여줍니다 . 다이어그램에서 10 -15  kmol 미만의 반응물 및 생성물은 표시되지 않았습니다. 이는 존재 하는 SF 6 의 양 (≈ 1.57 × 10 -10  kmol) 보다 5배 적 으므로 영향을 미치지 않습니다. 실제적인 방법으로 과정을 관찰했습니다.

그림 12

이 반응 과정은 3단계로 나눌 수 있다.

1단계 : 불화물의 형성. AZ91 용융물은 SF 6 및 그 분해 생성물과 우선적으로 반응하여 MgF 2 , AlF 3 및 ZnF 2 를 생성 합니다. 그러나 ZnF 2 의 양 이 너무 적어서 실제적으로 검출되지  않았을 수 있습니다(  MgF 2 의 3 × 10 -10 kmol에 비해 ZnF 2 1.25 × 10 -12 kmol ). 섹션 3.1 – 3.3에 표시된 모든 산화막 . 한편, 잔류 가스에 황이 SO 2 로 축적되었다 .

2단계 : 산화물의 형성. 액체 AZ91 합금이 포획된 가스에서 사용 가능한 모든 불화물을 고갈시킨 후, Mg와의 반응으로 인해 AlF 3 및 ZnF 2 의 양이 빠르게 감소했습니다. O 2 (g) 및 SO 2 는 AZ91 용융물과 반응하여 MgO, Al 2 O 3 , MgAl 2 O 4 , ZnO, ZnSO 4 및 MgSO 4 를 형성 합니다. 그러나 ZnO 및 ZnSO 4 의 양은 EDS에 의해 실제로 발견되기에는 너무 적었을 것입니다(예: 9.5 × 10 -12  kmol의 ZnO, 1.38 × 10 -14  kmol의 ZnSO 4 , 대조적으로 4.68 × 10−10  kmol의 MgF 2 , X 축의 AZ91 양 이 2.5 × 10 -9  kmol일 때). 실험 사례에서 커버 가스의 F 농도는 매우 낮고 전체 농도 f O는 훨씬 높습니다. 따라서 1단계와 2단계, 즉 불화물과 산화물의 형성은 반응 초기에 동시에 일어나 그림 1과 2와 같이 불화물과 산화물의 가수층 혼합물이 형성될 수 있다 . 4 및 10 (a). 내부 층은 산화물로 구성되어 있지만 불화물은 커버 가스에서 F 원소가 완전히 고갈된 후에 형성될 수 있습니다.

단계 1-2는 도 10 에 도시 된 다층 구조의 형성 과정을 이론적으로 검증하였다 .

산화막 내의 MgAl 2 O 4 및 Al 2 O 3 의 양은 도 4에 도시된 산화막과 일치하는 검출하기에 충분한 양이었다 . 그러나, 도 10 에 도시된 바와 같이, 산화셀에서 성장된 산화막에서는 알루미늄의 존재를 인식할 수 없었다 . 이러한 Al의 부재는 표면 필름과 AZ91 합금 용융물 사이의 다음 반응으로 인한 것일 수 있습니다.(1)

Al 2 O 3  + 3Mg + = 3MgO + 2Al, △G(700°C) = -119.82 kJ/mol(2)

Mg + MgAl 2 O 4  = MgO + Al, △G(700°C) = -106.34 kJ/mol이는 반응물이 서로 완전히 접촉한다는 가정 하에 열역학적 계산이 수행되었기 때문에 HSC 소프트웨어로 시뮬레이션할 수 없었습니다. 그러나 실제 공정에서 AZ91 용융물과 커버 가스는 보호 표면 필름의 존재로 인해 서로 완전히 접촉할 수 없습니다.

3단계 : 황화물과 질화물의 형성. 30분의 유지 시간 후, 산화 셀의 기상 불화물 및 산화물이 고갈되어 잔류 가스와 용융 반응을 허용하여 초기 F-농축 또는 (F, O )이 풍부한 표면 필름, 따라서 그림 10 (b 및 c)에 표시된 관찰된 다층 구조를 생성합니다 . 게다가, 질소는 모든 반응이 완료될 때까지 AZ91 용융물과 반응했습니다. 도 6 에 도시 된 산화막 은 질화물 함량으로 인해 이 반응 단계에 해당할 수 있다. 그러나, 그 결과는 도 1 및 도 5에 도시 된 연마된 샘플에서 질화물이 검출되지 않음을 보여준다. 4 와 5, 그러나 테스트 바 파단면에서만 발견됩니다. 질화물은 다음과 같이 샘플 준비 과정에서 가수분해될 수 있습니다 [54] .(삼)

Mg 3 N 2  + 6H 2 O = 3Mg(OH) 2  + 2NH 3 ↑(4)

AlN+ 3H 2 O = Al(OH) 3  + NH 3 ↑

또한 Schmidt et al. [61] 은 Mg 3 N 2 와 AlN이 반응하여 3원 질화물(Mg 3 Al n N n+2, n=1, 2, 3…) 을 형성할 수 있음을 발견했습니다 . HSC 소프트웨어에는 삼원 질화물 데이터베이스가 포함되어 있지 않아 계산에 추가할 수 없습니다. 이 단계의 산화막은 또한 삼원 질화물을 포함할 수 있습니다.

4.2 . SF 6 /CO 2 에서 형성된 연행 결함의 진화

도 13 은 AZ91 합금과 0.5%SF 6 /CO 2 사이의 열역학적 계산 결과를 보여준다 . 이 반응 과정도 세 단계로 나눌 수 있습니다.

그림 13

1단계 : 불화물의 형성. SF 6 및 그 분해 생성물은 AZ91 용융물에 의해 소비되어 MgF 2 , AlF 3 및 ZnF 2 를 형성했습니다 . 0.5% SF 6 /air 에서 AZ91의 반응에서와 같이 ZnF 2 의 양 이 너무 작아서 실제적으로 감지되지  않았습니다( 2.67 x 10 -10  kmol의 MgF 2 에 비해 ZnF 2 1.51 x 10 -13 kmol ). S와 같은 잔류 가스 트랩에 축적 유황 2 (g) 및 (S)의 일부분 (2) (g)가 CO와 반응하여 2 SO 형성하는 2및 CO. 이 반응 단계의 생성물은 도 11 (a)에 도시된 필름과 일치하며 , 이는 불화물만을 함유하는 단일 층 구조를 갖는다.

2단계 : 산화물의 형성. ALF 3 및 ZnF 2 MgF로 형성 용융 AZ91 마그네슘의 반응 2 , Al 및 Zn으로한다. SO 2 는 소모되기 시작하여 표면 필름에 산화물을 생성 하고 커버 가스에 S 2 (g)를 생성했습니다. 한편, CO 2 는 AZ91 용융물과 직접 반응하여 CO, MgO, ZnO 및 Al 2 O 3 를 형성 합니다. 도 1에 도시 된 산화막 9 및 11 (b)는 산소가 풍부한 층과 다층 구조로 인해 이 반응 단계에 해당할 수 있습니다.

커버 가스의 CO는 AZ91 용융물과 추가로 반응하여 C를 생성할 수 있습니다. 이 탄소는 온도가 감소할 때(응고 기간 동안) Mg와 추가로 반응하여 Mg 탄화물을 형성할 수 있습니다 [62] . 이것은 도 4에 도시된 산화막의 탄소 함량이 높은 이유일 수 있다 8 – 9 . Liang et al. [39] 또한 SO 2 /CO 2 로 보호된 AZ91 합금 표면 필름에서 탄소 검출을 보고했습니다 . 생성된 Al 2 O 3 는 MgO와 더 결합하여 MgAl 2 O [63]를 형성할 수 있습니다 . 섹션 4.1 에서 논의된 바와 같이, 알루미나 및 스피넬은 도 11 에 도시된 바와 같이 표면 필름에 알루미늄 부재를 야기하는 Mg와 반응할 수 있다 .

3단계 : 황화물의 형성. AZ91은 용융물 S 소비하기 시작 2 인 ZnS와 MGS 형성 갇힌 잔류 가스 (g)를. 이러한 반응은 반응 과정의 마지막 단계까지 일어나지 않았으며, 이는 Fig. 7 (c)에 나타난 결함의 S-함량 이 적은 이유일 수 있다 .

요약하면, 열역학적 계산은 AZ91 용융물이 커버 가스와 반응하여 먼저 불화물을 형성한 다음 마지막에 산화물과 황화물을 형성할 것임을 나타냅니다. 다른 반응 단계에서 산화막은 다른 구조와 조성을 가질 것입니다.

4.3 . 운반 가스가 동반 가스 소비 및 AZ91 주물의 재현성에 미치는 영향

SF 6 /air 및 SF 6 /CO 2 에서 형성된 연행 결함의 진화 과정은 4.1절 과 4.2  에서 제안되었습니다 . 이론적인 계산은 실제 샘플에서 발견되는 해당 산화막과 관련하여 검증되었습니다. 연행 결함 내의 대기는 Al-합금 시스템과 다른 시나리오에서 액체 Mg-합금과의 반응으로 인해 효율적으로 소모될 수 있습니다(즉, 연행된 기포의 질소가 Al-합금 용융물과 효율적으로 반응하지 않을 것입니다 [64 , 65] 그러나 일반적으로 “질소 연소”라고 하는 액체 Mg 합금에서 질소가 더 쉽게 소모될 것입니다 [66] ).

동반된 가스와 주변 액체 Mg-합금 사이의 반응은 동반된 가스를 산화막 내에서 고체 화합물(예: MgO)로 전환하여 동반 결함의 공극 부피를 감소시켜 결함(예: 공기의 동반된 가스가 주변의 액체 Mg 합금에 의해 고갈되면 용융 온도가 700 °C이고 액체 Mg 합금의 깊이가 10 cm라고 가정할 때 최종 고체 제품의 총 부피는 0.044가 됩니다. 갇힌 공기가 취한 초기 부피의 %).

연행 결함의 보이드 부피 감소와 해당 주조 특성 사이의 관계는 알루미늄 합금 주조에서 널리 연구되었습니다. Nyahumwa와 Campbell [16] 은 HIP(Hot Isostatic Pressing) 공정이 Al-합금 주물의 연행 결함이 붕괴되고 산화물 표면이 접촉하게 되었다고 보고했습니다. 주물의 피로 수명은 HIP 이후 개선되었습니다. Nyahumwa와 Campbell [16] 도 서로 접촉하고 있는 이중 산화막의 잠재적인 결합을 제안했지만 이를 뒷받침하는 직접적인 증거는 없었습니다. 이 결합 현상은 Aryafar et.al에 의해 추가로 조사되었습니다. [8], 그는 강철 튜브에서 산화물 스킨이 있는 두 개의 Al-합금 막대를 다시 녹인 다음 응고된 샘플에 대해 인장 강도 테스트를 수행했습니다. 그들은 Al-합금 봉의 산화물 스킨이 서로 강하게 결합되어 용융 유지 시간이 연장됨에 따라 더욱 강해짐을 발견했으며, 이는 이중 산화막 내 동반된 가스의 소비로 인한 잠재적인 “치유” 현상을 나타냅니다. 구조. 또한 Raidszadeh와 Griffiths [9 , 19] 는 연행 가스가 반응하는 데 더 긴 시간을 갖도록 함으로써 응고 전 용융 유지 시간을 연장함으로써 Al-합금 주물의 재현성에 대한 연행 결함의 부정적인 영향을 성공적으로 줄였습니다. 주변이 녹습니다.

앞서 언급한 연구를 고려할 때, Mg 합금 주물에서 혼입 가스의 소비는 다음 두 가지 방식으로 혼입 결함의 부정적인 영향을 감소시킬 수 있습니다.

(1) 이중 산화막의 결합 현상 . 도 5 및 도 7 에 도시 된 샌드위치형 구조 는 이중 산화막 구조의 잠재적인 결합을 나타내었다. 그러나 산화막의 결합으로 인한 강도 증가를 정량화하기 위해서는 더 많은 증거가 필요합니다.

(2) 연행 결함의 보이드 체적 감소 . 주조품의 품질에 대한 보이드 부피 감소의 긍정적인 효과는 HIP 프로세스 [67]에 의해 널리 입증되었습니다 . 섹션 4.1 – 4.2 에서 논의된 진화 과정과 같이 , 동반된 가스와 주변 AZ91 합금 용융물 사이의 지속적인 반응으로 인해 동반 결함의 산화막이 함께 성장할 수 있습니다. 최종 고체 생성물의 부피는 동반된 기체에 비해 상당히 작았다(즉, 이전에 언급된 바와 같이 0.044%).

따라서, 혼입 가스의 소모율(즉, 산화막의 성장 속도)은 AZ91 합금 주물의 품질을 향상시키는 중요한 매개변수가 될 수 있습니다. 이에 따라 산화 셀의 산화막 성장 속도를 추가로 조사했습니다.

도 14 는 상이한 커버 가스(즉, 0.5%SF 6 /air 및 0.5%SF 6 /CO 2 ) 에서의 표면 필름 성장 속도의 비교를 보여준다 . 필름 두께 측정을 위해 각 샘플의 15개의 임의 지점을 선택했습니다. 95% 신뢰구간(95%CI)은 막두께의 변화가 가우시안 분포를 따른다는 가정하에 계산하였다. 0.5%SF 6 /air 에서 형성된 모든 표면막이 0.5%SF 6 /CO 2 에서 형성된 것보다 빠르게 성장함을 알 수 있다 . 다른 성장률은 0.5%SF 6 /air 의 연행 가스 소비율 이 0.5%SF 6 /CO 2 보다 더 높음 을 시사했습니다., 이는 동반된 가스의 소비에 더 유리했습니다.

그림 14

산화 셀에서 액체 AZ91 합금과 커버 가스의 접촉 면적(즉, 도가니의 크기)은 많은 양의 용융물과 가스를 고려할 때 상대적으로 작았다는 점에 유의해야 합니다. 결과적으로, 산화 셀 내에서 산화막 성장을 위한 유지 시간은 비교적 길었다(즉, 5-30분). 하지만, 실제 주조에 함유 된 혼입 결함은 (상대적으로 매우 적은, 즉, 수 미크론의 크기에 도시 된 바와 같이 ,도 3. – 6 및 [7]), 동반된 가스는 주변 용융물로 완전히 둘러싸여 상대적으로 큰 접촉 영역을 생성합니다. 따라서 커버 가스와 AZ91 합금 용융물의 반응 시간은 비교적 짧을 수 있습니다. 또한 실제 Mg 합금 모래 주조의 응고 시간은 몇 분일 수 있습니다(예: Guo [68] 은 직경 60mm의 Mg 합금 모래 주조가 응고되는 데 4분이 필요하다고 보고했습니다). 따라서 Mg-합금 용융주조 과정에서 포획된 동반된 가스는 특히 응고 시간이 긴 모래 주물 및 대형 주물의 경우 주변 용융물에 의해 쉽게 소모될 것으로 예상할 수 있습니다.

따라서, 동반 가스의 다른 소비율과 관련된 다른 커버 가스(0.5%SF 6 /air 및 0.5%SF 6 /CO 2 )가 최종 주물의 재현성에 영향을 미칠 수 있습니다. 이 가정을 검증하기 위해 0.5%SF 6 /air 및 0.5%SF 6 /CO 2 에서 생산된 AZ91 주물 을 기계적 평가를 위해 테스트 막대로 가공했습니다. Weibull 분석은 선형 최소 자승(LLS) 방법과 비선형 최소 자승(비 LLS) 방법을 모두 사용하여 수행되었습니다 [69] .

그림 15 (ab)는 LLS 방법으로 얻은 UTS 및 AZ91 합금 주물의 연신율의 전통적인 2-p 선형 Weibull 플롯을 보여줍니다. 사용된 추정기는 P= (i-0.5)/N이며, 이는 모든 인기 있는 추정기 중 가장 낮은 편향을 유발하는 것으로 제안되었습니다 [69 , 70] . SF 6 /air 에서 생산된 주물 은 UTS Weibull 계수가 16.9이고 연신율 Weibull 계수가 5.0입니다. 대조적으로, SF 6 /CO 2 에서 생산된 주물의 UTS 및 연신 Weibull 계수는 각각 7.7과 2.7로, SF 6 /CO 2 에 의해 보호된 주물의 재현성이 SF 6 /air 에서 생산된 것보다 훨씬 낮음을 시사합니다. .

그림 15

또한 저자의 이전 출판물 [69] 은 선형화된 Weibull 플롯의 단점을 보여주었으며, 이는 Weibull 추정 의 더 높은 편향과 잘못된 2 중단을 유발할 수 있습니다 . 따라서 그림 15 (cd) 와 같이 Non-LLS Weibull 추정이 수행되었습니다 . SF 6 /공기주조물 의 UTS Weibull 계수 는 20.8인 반면, SF 6 /CO 2 하에서 생산된 주조물의 UTS Weibull 계수는 11.4로 낮아 재현성에서 분명한 차이를 보였다. 또한 SF 6 /air elongation(El%) 데이터 세트는 SF 6 /CO 2 의 elongation 데이터 세트보다 더 높은 Weibull 계수(모양 = 5.8)를 가졌습니다.(모양 = 3.1). 따라서 LLS 및 Non-LLS 추정 모두 SF 6 /공기 주조가 SF 6 /CO 2 주조 보다 더 높은 재현성을 갖는다고 제안했습니다 . CO 2 대신 공기를 사용 하면 혼입된 가스의 더 빠른 소비에 기여하여 결함 내의 공극 부피를 줄일 수 있다는 방법을 지원합니다 . 따라서 0.5%SF 6 /CO 2 대신 0.5%SF 6 /air를 사용 하면(동반된 가스의 소비율이 증가함) AZ91 주물의 재현성이 향상되었습니다.

그러나 모든 Mg 합금 주조 공장이 현재 작업에서 사용되는 주조 공정을 따랐던 것은 아니라는 점에 유의해야 합니다. Mg의 합금 용탕 본 작업은 탈기에 따라서, 동반 가스의 소비에 수소의 영향을 감소 (즉, 수소 잠재적 동반 가스의 고갈 억제, 동반 된 기체로 확산 될 수있다 [7 , 71 , 72] ). 대조적으로, 마그네슘 합금 주조 공장에서는 마그네슘을 주조할 때 ‘가스 문제’가 없고 따라서 인장 특성에 큰 변화가 없다고 널리 믿어지기 때문에 마그네슘 합금 용융물은 일반적으로 탈기되지 않습니다 [73] . 연구에 따르면 Mg 합금 주물의 기계적 특성에 대한 수소의 부정적인 영향 [41 ,42 , 73] , 탈기 공정은 마그네슘 합금 주조 공장에서 여전히 인기가 없습니다.

또한 현재 작업에서 모래 주형 공동은 붓기 전에 SF 6 커버 가스 로 플러싱되었습니다 [22] . 그러나 모든 Mg 합금 주조 공장이 이러한 방식으로 금형 캐비티를 플러싱한 것은 아닙니다. 예를 들어, Stone Foundry Ltd(영국)는 커버 가스 플러싱 대신 유황 분말을 사용했습니다. 그들의 주물 내의 동반된 가스 는 보호 가스라기 보다는 SO 2 /공기일 수 있습니다 .

따라서 본 연구의 결과는 CO 2 대신 공기를 사용 하는 것이 최종 주조의 재현성을 향상시키는 것으로 나타났지만 다른 산업용 Mg 합금 주조 공정과 관련하여 캐리어 가스의 영향을 확인하기 위해서는 여전히 추가 조사가 필요합니다.

7 . 결론

1.

AZ91 합금에 형성된 연행 결함이 관찰되었습니다. 그들의 산화막은 단층과 다층의 두 가지 유형의 구조를 가지고 있습니다. 다층 산화막은 함께 성장하여 최종 주조에서 샌드위치 같은 구조를 형성할 수 있습니다.2.

실험 결과와 이론적인 열역학적 계산은 모두 갇힌 가스의 불화물이 황을 소비하기 전에 고갈되었음을 보여주었습니다. 이중 산화막 결함의 3단계 진화 과정이 제안되었습니다. 산화막은 진화 단계에 따라 다양한 화합물 조합을 포함했습니다. SF 6 /air 에서 형성된 결함 은 SF 6 /CO 2 에서 형성된 것과 유사한 구조를 갖지만 산화막의 조성은 달랐다. 엔트레인먼트 결함의 산화막 형성 및 진화 과정은 이전에 보고된 Mg 합금 표면막(즉, MgF 2 이전에 형성된 MgO)의 것과 달랐다 .삼.

산화막의 성장 속도는 SF하에 큰 것으로 입증되었다 (6) / SF보다 공기 6 / CO 2 손상 봉입 가스의 빠른 소비에 기여한다. AZ91 합금 주물의 재현성은 SF 6 /CO 2 대신 SF 6 /air를 사용할 때 향상되었습니다 .

감사의 말

저자는 EPSRC LiME 보조금 EP/H026177/1의 자금 지원 과 WD Griffiths 박사와 Adrian Carden(버밍엄 대학교)의 도움을 인정합니다. 주조 작업은 University of Birmingham에서 수행되었습니다.

참조
[1]
MK McNutt , SALAZAR K.
마그네슘, 화합물 및 금속, 미국 지질 조사국 및 미국 내무부
레 스톤 , 버지니아 ( 2013 )
Google 학술검색
[2]
마그네슘
화합물 및 금속, 미국 지질 조사국 및 미국 내무부
( 1996 )
Google 학술검색
[삼]
I. Ostrovsky , Y. Henn
ASTEC’07 International Conference-New Challenges in Aeronautics , Moscow ( 2007 ) , pp. 1 – 5
8월 19-22일
Scopus에서 레코드 보기Google 학술검색
[4]
Y. Wan , B. Tang , Y. Gao , L. Tang , G. Sha , B. Zhang , N. Liang , C. Liu , S. Jiang , Z. Chen , X. Guo , Y. Zhao
액타 메이터. , 200 ( 2020 ) , 274 – 286 페이지
기사PDF 다운로드Scopus에서 레코드 보기
[5]
JTJ Burd , EA Moore , H. Ezzat , R. Kirchain , R. Roth
적용 에너지 , 283 ( 2021 ) , 제 116269 조
기사PDF 다운로드Scopus에서 레코드 보기
[6]
AM 루이스 , JC 켈리 , 조지아주 Keoleian
적용 에너지 , 126 ( 2014 ) , pp. 13 – 20
기사PDF 다운로드Scopus에서 레코드 보기
[7]
J. 캠벨
주물
버터워스-하이네만 , 옥스퍼드 ( 2004 )
Google 학술검색
[8]
M. Aryafar , R. Raiszadeh , A. Shalbafzadeh
J. 메이터. 과학. , 45 ( 2010 년 ) , PP. (3041) – 3051
교차 참조Scopus에서 레코드 보기
[9]
R. 라이자데 , WD 그리피스
메탈. 메이터. 트랜스. B-프로세스 메탈. 메이터. 프로세스. 과학. , 42 ( 2011 ) , 133 ~ 143페이지
교차 참조Scopus에서 레코드 보기
[10]
R. 라이자데 , WD 그리피스
J. 합금. Compd. , 491 ( 2010 ) , 575 ~ 580 쪽
기사PDF 다운로드Scopus에서 레코드 보기
[11]
L. Peng , G. Zeng , TC Su , H. Yasuda , K. Nogita , CM Gourlay
JOM , 71 ( 2019 ) , pp. 2235 – 2244
교차 참조Scopus에서 레코드 보기
[12]
S. Ganguly , AK Mondal , S. Sarkar , A. Basu , S. Kumar , C. Blawert
코로스. 과학. , 166 ( 2020 )
[13]
GE Bozchaloei , N. Varahram , P. Davami , SK 김
메이터. 과학. 영어 A-구조체. 메이터. 소품 Microstruct. 프로세스. , 548 ( 2012 ) , 99 ~ 105페이지
Scopus에서 레코드 보기
[14]
S. 폭스 , J. 캠벨
Scr. 메이터. , 43 ( 2000 ) , PP. 881 – 886
기사PDF 다운로드Scopus에서 레코드 보기
[15]
M. 콕스 , RA 하딩 , J. 캠벨
메이터. 과학. 기술. , 19 ( 2003 ) , 613 ~ 625페이지
Scopus에서 레코드 보기
[16]
C. Nyahumwa , NR Green , J. Campbell
메탈. 메이터. 트랜스. A-Phys. 메탈. 메이터. 과학. , 32 ( 2001 ) , 349 ~ 358 쪽
Scopus에서 레코드 보기
[17]
A. Ardekhani , R. Raiszadeh
J. 메이터. 영어 공연하다. , 21 ( 2012 ) , pp. 1352 – 1362
교차 참조Scopus에서 레코드 보기
[18]
X. Dai , X. Yang , J. Campbell , J. Wood
메이터. 과학. 기술. , 20 ( 2004 ) , 505 ~ 513 쪽
Scopus에서 레코드 보기
[19]
EM 엘갈라드 , MF 이브라힘 , HW 도티 , FH 사무엘
필로스. 잡지. , 98 ( 2018 ) , PP. 1337 – 1359
교차 참조Scopus에서 레코드 보기
[20]
WD 그리피스 , NW 라이
메탈. 메이터. 트랜스. A-Phys. 메탈. 메이터. 과학. , 38A ( 2007 ) , PP. 190 – 196
교차 참조Scopus에서 레코드 보기
[21]
AR Mirak , M. Divandari , SMA Boutorabi , J. 캠벨
국제 J. 캐스트 만났습니다. 해상도 , 20 ( 2007 ) , PP. 215 – 220
교차 참조Scopus에서 레코드 보기
[22]
C. 칭기
주조공학 연구실
Helsinki University of Technology , Espoo, Finland ( 2006 )
Google 학술검색
[23]
Y. Jia , J. Hou , H. Wang , Q. Le , Q. Lan , X. Chen , L. Bao
J. 메이터. 프로세스. 기술. , 278 ( 2020 ) , 제 116542 조
기사PDF 다운로드Scopus에서 레코드 보기
[24]
S. Ouyang , G. Yang , H. Qin , S. Luo , L. Xiao , W. Jie
메이터. 과학. 영어 A , 780 ( 2020 ) , 제 139138 조
기사PDF 다운로드Scopus에서 레코드 보기
[25]
에스엠. Xiong , X.-F. 왕
트랜스. 비철금속 사회 중국 , 20 ( 2010 ) , pp. 1228 – 1234
기사PDF 다운로드Scopus에서 레코드 보기
[26]
지브이리서치
그랜드뷰 리서치
( 2018 )
미국
Google 학술검색
[27]
T. 리 , J. 데이비스
메탈. 메이터. 트랜스. , 51 ( 2020 ) , PP. 5,389 – (5400)
교차 참조Scopus에서 레코드 보기
[28]
JF Fruehling, 미시간 대학, 1970.
Google 학술검색
[29]
S. 쿨링
제36회 세계 마그네슘 연례 회의 , 노르웨이 ( 1979 ) , pp. 54 – 57
Scopus에서 레코드 보기Google 학술검색
[30]
S. Cashion , N. Ricketts , P. Hayes
J. 가벼운 만남. , 2 ( 2002 ) , 43 ~ 47페이지
기사PDF 다운로드Scopus에서 레코드 보기
[31]
S. Cashion , N. Ricketts , P. Hayes
J. 가벼운 만남. , 2 ( 2002 ) , PP. 37 – 42
기사PDF 다운로드Scopus에서 레코드 보기
[32]
K. Aarstad , G. Tranell , G. Pettersen , TA Engh
SF6에 의해 보호되는 마그네슘의 표면을 연구하는 다양한 기술
TMS ( 2003년 )
Google 학술검색
[33]
에스엠 Xiong , X.-L. 리우
메탈. 메이터. 트랜스. , 38 ( 2007 년 ) , PP. (428) – (434)
교차 참조Scopus에서 레코드 보기
[34]
T.-S. 시 , J.-B. Liu , P.-S. 웨이
메이터. 화학 물리. , 104 ( 2007 ) , 497 ~ 504페이지
기사PDF 다운로드Scopus에서 레코드 보기
[35]
G. Pettersen , E. Øvrelid , G. Tranell , J. Fenstad , H. Gjestland
메이터. 과학. 영어 , 332 ( 2002 ) , PP. (285) – (294)
기사PDF 다운로드Scopus에서 레코드 보기
[36]
H. Bo , LB Liu , ZP Jin
J. 합금. Compd. , 490 ( 2010 ) , 318 ~ 325 쪽
기사PDF 다운로드Scopus에서 레코드 보기
[37]
A. 미락 , C. 데이비슨 , J. 테일러
코로스. 과학. , 52 ( 2010 ) , PP. 1992 년 – 2000
기사PDF 다운로드Scopus에서 레코드 보기
[38]
BD 리 , UH 부리 , KW 리 , GS 한강 , JW 한
메이터. 트랜스. , 54 ( 2013 ) , 66 ~ 73페이지
Scopus에서 레코드 보기
[39]
WZ Liang , Q. Gao , F. Chen , HH Liu , ZH Zhao
China Foundry , 9 ( 2012 ) , pp. 226 – 230
교차 참조Scopus에서 레코드 보기
[40]
UI 골드슐레거 , EY 샤피로비치
연소. 폭발 충격파 , 35 ( 1999 ) , 637 ~ 644페이지
Scopus에서 레코드 보기
[41]
A. Elsayed , SL Sin , E. Vandersluis , J. Hill , S. Ahmad , C. Ravindran , S. Amer Foundry
트랜스. 오전. 파운드리 Soc. , 120 ( 2012 ) , 423 ~ 429페이지
Scopus에서 레코드 보기
[42]
E. Zhang , GJ Wang , ZC Hu
메이터. 과학. 기술. , 26 ( 2010 ) , 1253 ~ 1258페이지
Scopus에서 레코드 보기
[43]
NR 그린 , J. 캠벨
메이터. 과학. 영어 A-구조체. 메이터. 소품 Microstruct. 프로세스. , 173 ( 1993 ) , 261 ~ 266 쪽
기사PDF 다운로드Scopus에서 레코드 보기
[44]
C 라일리 , MR 졸리 , NR 그린
MCWASP XII 논문집 – 주조, 용접 및 고급 Solidifcation 프로세스의 12 모델링 , 밴쿠버, 캐나다 ( 2009 )
Google 학술검색
[45]
HE Friedrich, BL Mordike, Springer, 독일, 2006.
Google 학술검색
[46]
C. Zheng , BR Qin , XB Lou
기계, 산업 및 제조 기술에 관한 2010 국제 회의 , ASME ( 2010 ) , pp. 383 – 388
2010년 미트
교차 참조Scopus에서 레코드 보기Google 학술검색
[47]
SM Xiong , XF 왕
트랜스. 비철금속 사회 중국 , 20 ( 2010 ) , pp. 1228 – 1234
기사PDF 다운로드Scopus에서 레코드 보기
[48]
SM Xiong , XL Liu
메탈. 메이터. 트랜스. A-Phys. 메탈. 메이터. 과학. , 38A ( 2007 ) , PP. (428) – (434)
교차 참조Scopus에서 레코드 보기
[49]
TS Shih , JB Liu , PS Wei
메이터. 화학 물리. , 104 ( 2007 ) , 497 ~ 504페이지
기사PDF 다운로드Scopus에서 레코드 보기
[50]
K. Aarstad , G. Tranell , G. Pettersen , TA Engh
매그. 기술. ( 2003 ) , PP. (5) – (10)
Scopus에서 레코드 보기
[51]
G. Pettersen , E. Ovrelid , G. Tranell , J. Fenstad , H. Gjestland
메이터. 과학. 영어 A-구조체. 메이터. 소품 Microstruct. 프로세스. , 332 ( 2002 ) , 285 ~ 294페이지
기사PDF 다운로드Scopus에서 레코드 보기
[52]
XF 왕 , SM Xiong
코로스. 과학. , 66 ( 2013 ) , PP. 300 – 307
기사PDF 다운로드Scopus에서 레코드 보기
[53]
SH Nie , SM Xiong , BC Liu
메이터. 과학. 영어 A-구조체. 메이터. 소품 Microstruct. 프로세스. , 422 ( 2006 ) , 346 ~ 351페이지
기사PDF 다운로드Scopus에서 레코드 보기
[54]
C. Bauer , A. Mogessie , U. Galovsky
Zeitschrift 모피 Metallkunde , 97 ( 2006 ) , PP. (164) – (168)
교차 참조Scopus에서 레코드 보기
[55]
QG 왕 , D. Apelian , DA Lados
J. 가벼운 만남. , 1 ( 2001 ) , PP. (73) – 84
기사PDF 다운로드Scopus에서 레코드 보기
[56]
S. Wang , Y. Wang , Q. Ramasse , Z. Fan
메탈. 메이터. 트랜스. , 51 ( 2020 ) , PP. 2957 – 2974
교차 참조Scopus에서 레코드 보기
[57]
S. Hayashi , W. Minami , T. Oguchi , HJ Kim
카그. 코그. 론분슈 , 35 ( 2009 ) , 411 ~ 415페이지
교차 참조Scopus에서 레코드 보기
[58]
K. 아르스타드
노르웨이 과학 기술 대학교
( 2004년 )
Google 학술검색
[59]
RL 윌킨스
J. Chem. 물리. , 51 ( 1969 ) , p. 853
-&
Scopus에서 레코드 보기
[60]
O. Kubaschewski , K. Hesselemam
무기물의 열화학적 성질
Springer-Verlag , 벨린 ( 1991 )
Google 학술검색
[61]
R. Schmidt , M. Strobele , K. Eichele , HJ Meyer
유로 J. Inorg. 화학 ( 2017 ) , PP. 2727 – 2735
교차 참조Scopus에서 레코드 보기
[62]
B. Hu , Y. Du , H. Xu , W. Sun , WW Zhang , D. Zhao
제이민 메탈. 분파. B-금속. , 46 ( 2010 ) , 97 ~ 103페이지
Scopus에서 레코드 보기
[63]
O. Salas , H. Ni , V. Jayaram , KC Vlach , CG Levi , R. Mehrabian
J. 메이터. 해상도 , 6 ( 1991 ) , 1964 ~ 1981페이지
Scopus에서 레코드 보기
[64]
SSS Kumari , UTS Pillai , BC 빠이
J. 합금. Compd. , 509 ( 2011 ) , pp. 2503 – 2509
기사PDF 다운로드Scopus에서 레코드 보기
[65]
H. Scholz , P. Greil
J. 메이터. 과학. , 26 ( 1991 ) , 669 ~ 677 쪽
Scopus에서 레코드 보기
[66]
P. Biedenkopf , A. Karger , M. Laukotter , W. Schneider
매그. 기술. , 2005년 ( 2005년 ) , 39 ~ 42 쪽
Scopus에서 레코드 보기
[67]
HV 앳킨슨 , S. 데이비스
메탈. 메이터. 트랜스. , 31 ( 2000 ) , PP. 2981 – 3000
교차 참조Scopus에서 레코드 보기
[68]
EJ Guo , L. Wang , YC Feng , LP Wang , YH Chen
J. 썸. 항문. 칼로리. , 135 ( 2019 ) , PP. 2001 년 – 2008 년
교차 참조Scopus에서 레코드 보기
[69]
T. Li , WD Griffiths , J. Chen
메탈. 메이터. 트랜스. A-Phys. 메탈. 메이터. 과학. , 48A ( 2017 ) , PP. 5516 – 5528
교차 참조Scopus에서 레코드 보기
[70]
M. Tiryakioglu , D. Hudak는
J. 메이터. 과학. , 42 ( 2007 ) , pp. 10173 – 10179
교차 참조Scopus에서 레코드 보기
[71]
Y. Yue , WD Griffiths , JL Fife , NR Green
제1회 3d 재료과학 국제학술대회 논문집 ( 2012 ) , pp. 131 – 136
교차 참조Scopus에서 레코드 보기Google 학술검색
[72]
R. 라이자데 , WD 그리피스
메탈. 메이터. 트랜스. B-프로세스 메탈. 메이터. 프로세스. 과학. , 37 ( 2006 ) , PP. (865) – (871)
Scopus에서 레코드 보기
[73]
ZC Hu , EL Zhang , SY Zeng
메이터. 과학. 기술. , 24 ( 2008 ) , 1304 ~ 1308페이지
교차 참조Scopus에서 레코드 보기

Conflict resolution in the multi-stakeholder stepped spillway design under uncertainty by machine learning techniques

기계 학습 기술에 의한 불확실성 하에서 다중 이해 관계자 계단형 배수로 설계의 충돌 해결

Conflict resolution in the multi-stakeholder stepped spillway design under uncertainty by machine learning techniques

Mehrdad GhorbaniMooseluaMohammad RezaNikoobParnian HashempourBakhtiaribNooshin BakhtiariRayanicAzizallahIzadyd
aDepartment of Engineering Sciences, University of Agder, Norway
bDepartment of Civil and Environmental Engineering, Shiraz University, Shiraz, Iran
cSchool of Engineering, Department of Civil and Environmental Engineering, Shiraz University, Shiraz, IrandWater Research Center, Sultan Qaboos University, Muscat, Oman

Abstract

The optimal spillway design is of great significance since these structures can reduce erosion downstream of the dams. This study proposes a risk-based optimization framework for a stepped spillway to achieve an economical design scenario with the minimum loss in hydraulic performance. Accordingly, the stepped spillway was simulated in the FLOW-3D® model, and the validated model was repeatedly performed for various geometric states.

The results were used to form a Multilayer Perceptron artificial neural network (MLP-ANN) surrogate model. Then, a risk-based optimization model was formed by coupling the MLP-ANN and NSGA-II. The concept of conditional value at risk (CVaR) was utilized to reduce the risk of the designed spillway malfunctions in high flood flow rates, while minimizing the construction cost and the loss in hydraulic performance.

Lastly, given the conflicting objectives of stakeholders, the non-cooperative graph model for conflict resolution (GMCR) was applied to achieve a compromise on the Pareto optimal solutions. Applicability of the suggested approach in the Jarreh Dam, Iran, resulted in a practical design scenario, which simultaneously minimizes the loss in hydraulic performance and the project cost and satisfies the priorities of decision-makers.

Keywords

Stepped spillway, FLOW-3D® ,CVaR-based optimization model, GMCR-plus, NSGA-II

최적의 배수로 설계는 이러한 구조가 댐 하류의 침식을 줄일 수 있기 때문에 매우 중요합니다. 본 연구에서는 유압 성능 손실을 최소화하면서 경제적인 설계 시나리오를 달성하기 위해 계단형 여수로에 대한 위험 기반 최적화 프레임워크를 제안합니다. 따라서 FLOW-3D® 모델에서 계단식 배수로를 시뮬레이션하고 다양한 기하학적 상태에 대해 검증된 모델을 반복적으로 수행했습니다.

결과는 다층 퍼셉트론 인공 신경망(MLP-ANN) 대리 모델을 형성하는 데 사용되었습니다. 그런 다음 MLP-ANN과 NSGA-II를 결합하여 위험 기반 최적화 모델을 구성했습니다. 위험 조건부 값(CVaR)의 개념은 높은 홍수 유량에서 설계된 방수로 오작동의 위험을 줄이는 동시에 건설 비용과 수리 성능 손실을 최소화하기 위해 활용되었습니다.

마지막으로 이해관계자의 상충되는 목표를 고려하여 파레토 최적해에 대한 절충안을 달성하기 위해 갈등 해결을 위한 비협조적 그래프 모델(GMCR)을 적용하였다. 이란 Jarreh 댐에서 제안된 접근 방식의 적용 가능성은 수력 성능 손실과 프로젝트 비용을 동시에 최소화하고 의사 결정자의 우선 순위를 만족시키는 실용적인 설계 시나리오로 귀결되었습니다.

FLOW-3D AM

flow3d AM-product
FLOW-3D AM-product

와이어 파우더 기반 DED | Wire Powder Based DED

일부 연구자들은 부품을 만들기 위해 더 넓은 범위의 처리 조건을 사용하여 하이브리드 와이어 분말 기반 DED 시스템을 찾고 있습니다. 예를 들어, 이 시뮬레이션은 다양한 분말 및 와이어 이송 속도를 가진 하이브리드 시스템을 살펴봅니다.

와이어 기반 DED | Wire Based DED

와이어 기반 DED는 분말 기반 DED보다 처리량이 높고 낭비가 적지만 재료 구성 및 증착 방향 측면에서 유연성이 떨어집니다. FLOW-3D AM 은 와이어 기반 DED의 처리 결과를 이해하는데 유용하며 최적화 연구를 통해 빌드에 대한 와이어 이송 속도 및 직경과 같은 최상의 처리 매개 변수를 찾을 수 있습니다.

FLOW-3D AM은 레이저 파우더 베드 융합 (L-PBF), 바인더 제트 및 DED (Directed Energy Deposition)와 같은 적층 제조 공정 ( additive manufacturing )을 시뮬레이션하고 분석하는 CFD 소프트웨어입니다. FLOW-3D AM 의 다중 물리 기능은 공정 매개 변수의 분석 및 최적화를 위해 분말 확산 및 압축, 용융 풀 역학, L-PBF 및 DED에 대한 다공성 형성, 바인더 분사 공정을 위한 수지 침투 및 확산에 대해 매우 정확한 시뮬레이션을 제공합니다.

3D 프린팅이라고도하는 적층 제조(additive manufacturing)는 일반적으로 층별 접근 방식을 사용하여, 분말 또는 와이어로 부품을 제조하는 방법입니다. 금속 기반 적층 제조 공정에 대한 관심은 지난 몇 년 동안 시작되었습니다. 오늘날 사용되는 3 대 금속 적층 제조 공정은 PBF (Powder Bed Fusion), DED (Directed Energy Deposition) 및 바인더 제트 ( Binder jetting ) 공정입니다.  FLOW-3D  AM  은 이러한 각 프로세스에 대한 고유 한 시뮬레이션 통찰력을 제공합니다.

파우더 베드 융합 및 직접 에너지 증착 공정에서 레이저 또는 전자 빔을 열원으로 사용할 수 있습니다. 두 경우 모두 PBF용 분말 형태와 DED 공정용 분말 또는 와이어 형태의 금속을 완전히 녹여 융합하여 층별로 부품을 형성합니다. 그러나 바인더 젯팅(Binder jetting)에서는 결합제 역할을 하는 수지가 금속 분말에 선택적으로 증착되어 층별로 부품을 형성합니다. 이러한 부품은 더 나은 치밀화를 달성하기 위해 소결됩니다.

FLOW-3D AM 의 자유 표면 추적 알고리즘과 다중 물리 모델은 이러한 각 프로세스를 높은 정확도로 시뮬레이션 할 수 있습니다. 레이저 파우더 베드 융합 (L-PBF) 공정 모델링 단계는 여기에서 자세히 설명합니다. DED 및 바인더 분사 공정에 대한 몇 가지 개념 증명 시뮬레이션도 표시됩니다.

레이저 파우더 베드 퓨전 (L-PBF)

LPBF 공정에는 유체 흐름, 열 전달, 표면 장력, 상 변화 및 응고와 같은 복잡한 다중 물리학 현상이 포함되어 공정 및 궁극적으로 빌드 품질에 상당한 영향을 미칩니다. FLOW-3D AM 의 물리적 모델은 질량, 운동량 및 에너지 보존 방정식을 동시에 해결하는 동시에 입자 크기 분포 및 패킹 비율을 고려하여 중규모에서 용융 풀 현상을 시뮬레이션합니다.

FLOW-3D DEM FLOW-3D WELD 는 전체 파우더 베드 융합 공정을 시뮬레이션하는 데 사용됩니다. L-PBF 공정의 다양한 단계는 분말 베드 놓기, 분말 용융 및 응고,이어서 이전에 응고 된 층에 신선한 분말을 놓는 것, 그리고 다시 한번 새 층을 이전 층에 녹이고 융합시키는 것입니다. FLOW-3D AM  은 이러한 각 단계를 시뮬레이션하는 데 사용할 수 있습니다.

파우더 베드 부설 공정

FLOW-3D DEM을 통해 분말 크기 분포, 재료 특성, 응집 효과는 물론 롤러 또는 블레이드 움직임 및 상호 작용과 같은 기하학적 효과와 관련된 분말 확산 및 압축을 이해할 수 있습니다. 이러한 시뮬레이션은 공정 매개 변수가 후속 인쇄 공정에서 용융 풀 역학에 직접적인 영향을 미치는 패킹 밀도와 같은 분말 베드 특성에 어떻게 영향을 미치는지에 대한 정확한 이해를 제공합니다.

다양한 파우더 베드 압축을 달성하는 한 가지 방법은 베드를 놓는 동안 다양한 입자 크기 분포를 선택하는 것입니다. 아래에서 볼 수 있듯이 세 가지 크기의 입자 크기 분포가 있으며, 이는 가장 높은 압축을 제공하는 Case 2와 함께 다양한 분말 베드 압축을 초래합니다.

파우더 베드 분포 다양한 입자 크기 분포
세 가지 다른 입자 크기 분포를 사용하여 파우더 베드 배치
파우더 베드 압축 결과
세 가지 다른 입자 크기 분포를 사용한 분말 베드 압축

입자-입자 상호 작용, 유체-입자 결합 및 입자 이동 물체 상호 작용은 FLOW-3D DEM을 사용하여 자세히 분석 할 수도 있습니다 . 또한 입자간 힘을 지정하여 분말 살포 응용 분야를 보다 정확하게 연구 할 수도 있습니다.

FLOW-3D AM  시뮬레이션은 이산 요소 방법 (DEM)을 사용하여 역 회전하는 원통형 롤러로 인한 분말 확산을 연구합니다. 비디오 시작 부분에서 빌드 플랫폼이 위로 이동하는 동안 분말 저장소가 아래로 이동합니다. 그 직후, 롤러는 분말 입자 (초기 위치에 따라 색상이 지정됨)를 다음 층이 녹고 구축 될 준비를 위해 구축 플랫폼으로 펼칩니다. 이러한 시뮬레이션은 저장소에서 빌드 플랫폼으로 전송되는 분말 입자의 선호 크기에 대한 추가 통찰력을 제공 할 수 있습니다.

Melting | 파우더 베드 용해

DEM 시뮬레이션에서 파우더 베드가 생성되면 STL 파일로 추출됩니다. 다음 단계는 CFD를 사용하여 레이저 용융 공정을 시뮬레이션하는 것입니다. 여기서는 레이저 빔과 파우더 베드의 상호 작용을 모델링 합니다. 이 프로세스를 정확하게 포착하기 위해 물리학에는 점성 흐름, 용융 풀 내의 레이저 반사 (광선 추적을 통해), 열 전달, 응고, 상 변화 및 기화, 반동 압력, 차폐 가스 압력 및 표면 장력이 포함됩니다. 이 모든 물리학은 이 복잡한 프로세스를 정확하게 시뮬레이션하기 위해 TruVOF 방법을 기반으로 개발되었습니다.

레이저 출력 200W, 스캔 속도 3.0m / s, 스폿 반경 100μm에서 파우더 베드의 용융 풀 분석.

용융 풀이 응고되면 FLOW-3D AM  압력 및 온도 데이터를 Abaqus 또는 MSC Nastran과 같은 FEA 도구로 가져와 응력 윤곽 및 변위 프로파일을 분석 할 수도 있습니다.

Multilayer | 다층 적층 제조

용융 풀 트랙이 응고되면 DEM을 사용하여 이전에 응고된 층에 새로운 분말 층의 확산을 시뮬레이션 할 수 있습니다. 유사하게, 레이저 용융은 새로운 분말 층에서 수행되어 후속 층 간의 융합 조건을 분석 할 수 있습니다.

해석 진행 절차는 첫 번째 용융층이 응고되면 입자의 두 번째 층이 응고 층에 증착됩니다. 새로운 분말 입자 층에 레이저 공정 매개 변수를 지정하여 용융 풀 시뮬레이션을 다시 수행합니다. 이 프로세스를 여러 번 반복하여 연속적으로 응고된 층 간의 융합, 빌드 내 온도 구배를 평가하는 동시에 다공성 또는 기타 결함의 형성을 모니터링 할 수 있습니다.

다층 적층 적층 제조 시뮬레이션

LPBF의 키홀 링 | Keyholing in LPBF

키홀링 중 다공성은 어떻게 형성됩니까? 이것은 TU Denmark의 연구원들이 FLOW-3D AM을 사용하여 답변한 질문이었습니다. 레이저 빔의 적용으로 기판이 녹으면 기화 및 상 변화로 인한 반동 압력이 용융 풀을 압박합니다. 반동 압력으로 인한 하향 흐름과 레이저 반사로 인한 추가 레이저 에너지 흡수가 공존하면 폭주 효과가 발생하여 용융 풀이 Keyholing으로 전환됩니다. 결국, 키홀 벽을 따라 온도가 변하기 때문에 표면 장력으로 인해 벽이 뭉쳐져서 진행되는 응고 전선에 의해 갇힐 수 있는 공극이 생겨 다공성이 발생합니다. FLOW-3D AM 레이저 파우더 베드 융합 공정 모듈은 키홀링 및 다공성 형성을 시뮬레이션 하는데 필요한 모든 물리 모델을 보유하고 있습니다.

바인더 분사 (Binder jetting)

Binder jetting 시뮬레이션은 모세관 힘의 영향을받는 파우더 베드에서 바인더의 확산 및 침투에 대한 통찰력을 제공합니다. 공정 매개 변수와 재료 특성은 증착 및 확산 공정에 직접적인 영향을 미칩니다.

Scan Strategy | 스캔 전략

스캔 전략은 온도 구배 및 냉각 속도에 영향을 미치기 때문에 미세 구조에 직접적인 영향을 미칩니다. 연구원들은 FLOW-3D AM 을 사용하여 결함 형성과 응고된 금속의 미세 구조에 영향을 줄 수 있는 트랙 사이에서 발생하는 재 용융을 이해하기 위한 최적의 스캔 전략을 탐색하고 있습니다. FLOW-3D AM 은 하나 또는 여러 레이저에 대해 시간에 따른 방향 속도를 구현할 때 완전한 유연성을 제공합니다.

Beam Shaping | 빔 형성

레이저 출력 및 스캔 전략 외에도 레이저 빔 모양과 열유속 분포는 LPBF 공정에서 용융 풀 역학에 큰 영향을 미칩니다. AM 기계 제조업체는 공정 안정성 및 처리량에 대해 다중 코어 및 임의 모양의 레이저 빔 사용을 모색하고 있습니다. FLOW-3D AM을 사용하면 멀티 코어 및 임의 모양의 빔 프로파일을 구현할 수 있으므로 생산량을 늘리고 부품 품질을 개선하기 위한 최상의 구성에 대한 통찰력을 제공 할 수 있습니다.

이 영역에서 수행 된 일부 작업에 대해 자세히 알아 보려면 “The Next Frontier of Metal AM”웨비나를 시청하십시오.

Multi-material Powder Bed Fusion | 다중 재료 분말 베드 융합

이 시뮬레이션에서 스테인리스 강 및 알루미늄 분말은 FLOW-3D AM 이 용융 풀 역학을 정확하게 포착하기 위해 추적하는 독립적으로 정의 된 온도 의존 재료 특성을 가지고 있습니다. 시뮬레이션은 용융 풀에서 재료 혼합을 이해하는 데 도움이됩니다.

다중 재료 용접 사례 연구

이종 금속의 레이저 키홀 용접에서 금속 혼합 조사

GM과 University of Utah의 연구원들은 FLOW-3D WELD 를 사용 하여 레이저 키홀 용접을 통한 이종 금속의 혼합을 이해했습니다. 그들은 반동 압력 및 Marangoni 대류와 관련하여 구리와 알루미늄의 혼합 농도에 대한 레이저 출력 및 스캔 속도의 영향을 조사했습니다. 그들은 시뮬레이션을 실험 결과와 비교했으며 샘플 내의 절단 단면에서 재료 농도 사이에 좋은 일치를 발견했습니다.

이종 금속의 레이저 키홀 용접에서 금속 혼합 조사
이종 금속의 레이저 키홀 용접에서 금속 혼합 조사
참조 : Wenkang Huang, Hongliang Wang, Teresa Rinker, Wenda Tan, 이종 금속의 레이저 키홀 용접에서 금속 혼합 조사 , Materials & Design, Volume 195, (2020). https://doi.org/10.1016/j.matdes.2020.109056
참조 : Wenkang Huang, Hongliang Wang, Teresa Rinker, Wenda Tan, 이종 금속의 레이저 키홀 용접에서 금속 혼합 조사 , Materials & Design, Volume 195, (2020). https://doi.org/10.1016/j.matdes.2020.109056

방향성 에너지 증착

FLOW-3D AM 의 내장 입자 모델 을 사용하여 직접 에너지 증착 프로세스를 시뮬레이션 할 수 있습니다. 분말 주입 속도와 고체 기질에 입사되는 열유속을 지정함으로써 고체 입자는 용융 풀에 질량, 운동량 및 에너지를 추가 할 수 있습니다. 다음 비디오에서 고체 금속 입자가 용융 풀에 주입되고 기판에서 용융 풀의 후속 응고가 관찰됩니다.

Figure 6. Evolution of melt pool in the overhang region (θ = 45°, P = 100 W, v = 1000 mm/s, the streamlines are shown by arrows).

Experimental and numerical investigation of the origin of surface roughness in laser powder bed fused overhang regions

레이저 파우더 베드 융합 오버행 영역에서 표면 거칠기의 원인에 대한 실험 및 수치 조사

Shaochuan Feng,Amar M. Kamat,Soheil Sabooni &Yutao PeiPages S66-S84 | Received 18 Jan 2021, Accepted 25 Feb 2021, Published online: 10 Mar 2021

ABSTRACT

Surface roughness of laser powder bed fusion (L-PBF) printed overhang regions is a major contributor to deteriorated shape accuracy/surface quality. This study investigates the mechanisms behind the evolution of surface roughness (Ra) in overhang regions. The evolution of surface morphology is the result of a combination of border track contour, powder adhesion, warp deformation, and dross formation, which is strongly related to the overhang angle (θ). When 0° ≤ θ ≤ 15°, the overhang angle does not affect Ra significantly since only a small area of the melt pool boundaries contacts the powder bed resulting in slight powder adhesion. When 15° < θ ≤ 50°, powder adhesion is enhanced by the melt pool sinking and the increased contact area between the melt pool boundary and powder bed. When θ > 50°, large waviness of the overhang contour, adhesion of powder clusters, severe warp deformation and dross formation increase Ra sharply.

레이저 파우더 베드 퓨전 (L-PBF) 프린팅 오버행 영역의 표면 거칠기는 형상 정확도 / 표면 품질 저하의 주요 원인입니다. 이 연구 는 오버행 영역에서 표면 거칠기 (Ra ) 의 진화 뒤에 있는 메커니즘을 조사합니다 . 표면 형태의 진화는 오버행 각도 ( θ ) 와 밀접한 관련이있는 경계 트랙 윤곽, 분말 접착, 뒤틀림 변형 및 드로스 형성의 조합의 결과입니다 . 0° ≤  θ  ≤ 15° 인 경우 , 용융풀 경계의 작은 영역 만 분말 베드와 접촉하여 약간의 분말 접착이 발생하기 때문에 오버행 각도가 R a에 큰 영향을 주지 않습니다 . 15° < θ 일 때  ≤ 50°, 용융 풀 싱킹 및 용융 풀 경계와 분말 베드 사이의 증가된 접촉 면적으로 분말 접착력이 향상됩니다. θ  > 50° 일 때 오버행 윤곽의 큰 파형, 분말 클러스터의 접착, 심한 휨 변형 및 드 로스 형성이 Ra 급격히 증가 합니다.

KEYWORDS: Laser powder bed fusion (L-PBF), melt pool dynamics, overhang region, shape deviation, surface roughness

1. Introduction

레이저 분말 베드 융합 (L-PBF)은 첨단 적층 제조 (AM) 기술로, 집중된 레이저 빔을 사용하여 금속 분말을 선택적으로 융합하여 슬라이스 된 3D 컴퓨터 지원에 따라 층별로 3 차원 (3D) 금속 부품을 구축합니다. 설계 (CAD) 모델 (Chatham, Long 및 Williams 2019 ; Tan, Zhu 및 Zhou 2020 ). 재료가 인쇄 층 아래에 ​​존재하는지 여부에 따라 인쇄 영역은 각각 솔리드 영역 또는 돌출 영역으로 분류 될 수 있습니다. 따라서 오버행 영역은 고체 기판이 아니라 분말 베드 바로 위에 건설되는 특수 구조입니다 (Patterson, Messimer 및 Farrington 2017). 오버행 영역은지지 구조를 포함하거나 포함하지 않고 구축 할 수 있으며, 지지대가있는 돌출 영역의 L-PBF는 지지체가 더 낮은 밀도로 구축된다는 점을 제외 하고 (Wang and Chou 2018 ) 고체 기판의 공정과 유사합니다 (따라서 기계적 강도가 낮기 때문에 L-PBF 공정 후 기계적으로 쉽게 제거 할 수 있습니다. 따라서지지 구조로 인쇄 된 오버행 영역은 L-PBF 공정 후 지지물 제거, 연삭 및 연마와 같은 추가 후 처리 단계가 필요합니다.

수평 내부 채널의 제작과 같은 일부 특정 경우에는 공정 후 지지대를 제거하기가 어려우므로 채널 상단 절반의 돌출부 영역을 지지대없이 건설해야합니다 (Hopkinson and Dickens 2000 ). 수평 내부 채널에 사용할 수없는지지 구조 외에도 내부 표면, 특히 등각 냉각 채널 (Feng, Kamat 및 Pei 2021 ) 에서 발생하는 복잡한 3D 채널 네트워크의 경우 표면 마감 프로세스를 구현하는 것도 어렵습니다 . 결과적으로 오버행 영역은 (i) 잔류 응력에 의한 변형, (ii) 계단 효과 (Kuo et al. 2020 ; Li et al. 2020 )로 인해 설계된 모양에서 벗어날 수 있습니다 .) 및 (iii) 원하지 않는 분말 소결로 인한 향상된 표면 거칠기; 여기서, 앞의 두 요소는 일반적으로 mm 길이 스케일에서 ‘매크로’편차로 분류되고 후자는 일반적으로 µm 길이 스케일에서 ‘마이크로’편차로 인식됩니다.

열 응력에 의한 변형은 오버행 영역에서 발생하는 중요한 문제입니다 (Patterson, Messimer 및 Farrington 2017 ). 국부적 인 용융 / 냉각은 용융 풀 내부 및 주변에서 큰 온도 구배를 유도하여 응고 된 층에 집중적 인 열 응력을 유발합니다. 열 응력에 의한 뒤틀림은 고체 영역을 현저하게 변형하지 않습니다. 이러한 영역은 아래의 여러 레이어에 의해 제한되기 때문입니다. 반면에 오버행 영역은 구속되지 않고 공정 중 응력 완화로 인해 상당한 변형이 발생합니다 (Kamat 및 Pei 2019 ). 더욱이 용융 깊이는 레이어 두께보다 큽니다 (이전 레이어도 재용 해되어 빌드 된 레이어간에 충분한 결합을 보장하기 때문입니다 [Yadroitsev et al. 2013 ; Kamath et al.2014 ]),응고 된 두께가 설계된 두께보다 크기 때문에형태 편차 (예 : 드 로스 [Charles et al. 2020 ; Feng et al. 2020 ])가 발생합니다. 마이크로 스케일에서 인쇄 된 표면 (R a 및 S a ∼ 10 μm)은 기계적으로 가공 된 표면보다 거칠다 (Duval-Chaneac et al. 2018 ; Wen et al. 2018 ). 이 문제는고형화 된 용융 풀의 가장자리에 부착 된 용융되지 않은 분말의 결과로 표면 거칠기 (R a )가 일반적으로 약 20 μm인 오버행 영역에서 특히 심각합니다 (Mazur et al. 2016 ; Pakkanen et al. 2016 ).

오버행 각도 ( θ , 빌드 방향과 관련하여 측정)는 오버행 영역의 뒤틀림 편향과 표면 거칠기에 영향을 미치는 중요한 매개 변수입니다 (Kamat and Pei 2019 ; Mingear et al. 2019 ). θ ∼ 45 ° 의 오버행 각도 는 일반적으로지지 구조없이 오버행 영역을 인쇄 할 수있는 임계 값으로 합의됩니다 (Pakkanen et al. 2016 ; Kadirgama et al. 2018 ). θ 일 때이 임계 값보다 크면 오버행 영역을 허용 가능한 표면 품질로 인쇄 할 수 없습니다. 오버행 각도 외에도 레이저 매개 변수 (레이저 에너지 밀도와 관련된)는 용융 풀의 모양 / 크기 및 용융 풀 역학에 영향을줌으로써 오버행 영역의 표면 거칠기에 영향을줍니다 (Wang et al. 2013 ; Mingear et al . 2019 ).

용융 풀 역학은 고체 (Shrestha 및 Chou 2018 ) 및 오버행 (Le et al. 2020 ) 영역 모두에서 수행되는 L-PBF 공정을 포함한 레이저 재료 가공의 일반적인 물리적 현상입니다 . 용융 풀 모양, 크기 및 냉각 속도는 잔류 응력으로 인한 변형과 ​​표면 거칠기에 모두 영향을 미치므로 처리 매개 변수와 표면 형태 / 품질 사이의 다리 역할을하며 용융 풀을 이해하기 위해 수치 시뮬레이션을 사용하여 추가 조사를 수행 할 수 있습니다. 거동과 표면 거칠기에 미치는 영향. 현재까지 고체 영역의 L-PBF 동안 용융 풀 동작을 시뮬레이션하기 위해 여러 연구가 수행되었습니다. 유한 요소 방법 (FEM)과 같은 시뮬레이션 기술 (Roberts et al. 2009 ; Du et al.2019 ), 유한 차분 법 (FDM) (Wu et al. 2018 ), 전산 유체 역학 (CFD) (Lee and Zhang 2016 ), 임의의 Lagrangian-Eulerian 방법 (ALE) (Khairallah and Anderson 2014 )을 사용하여 증발 반동 압력 (Hu et al. 2018 ) 및 Marangoni 대류 (Zhang et al. 2018 ) 현상을포함하는 열 전달 (온도 장) 및 물질 전달 (용융 흐름) 프로세스. 또한 이산 요소법 (DEM)을 사용하여 무작위 분산 분말 베드를 생성했습니다 (Lee and Zhang 2016 ; Wu et al. 2018 ). 이 모델은 분말 규모의 L-PBF 공정을 시뮬레이션했습니다 (Khairallah et al. 2016) 메조 스케일 (Khairallah 및 Anderson 2014 ), 단일 트랙 (Leitz et al. 2017 )에서 다중 트랙 (Foroozmehr et al. 2016 ) 및 다중 레이어 (Huang, Khamesee 및 Toyserkani 2019 )로.

그러나 결과적인 표면 거칠기를 결정하는 오버행 영역의 용융 풀 역학은 문헌에서 거의 관심을받지 못했습니다. 솔리드 영역의 L-PBF에 대한 기존 시뮬레이션 모델이 어느 정도 참조가 될 수 있지만 오버행 영역과 솔리드 영역 간의 용융 풀 역학에는 상당한 차이가 있습니다. 오버행 영역에서 용융 금속은 분말 입자 사이의 틈새로 아래로 흘러 용융 풀이 다공성 분말 베드가 제공하는 약한 지지체 아래로 가라 앉습니다. 이것은 중력과 표면 장력의 영향이 용융 풀의 결과적인 모양 / 크기를 결정하는 데 중요하며, 결과적으로 오버행 영역의 마이크로 스케일 형태의 진화에 중요합니다. 또한 분말 입자 사이의 공극, 열 조건 (예 : 에너지 흡수,2019 ; Karimi et al. 2020 ; 노래와 영 2020 ). 표면 거칠기는 (마이크로) 형상 편차를 증가시킬뿐만 아니라 주기적 하중 동안 미세 균열의 시작 지점 역할을함으로써 기계적 강도를 저하시킵니다 (Günther et al. 2018 ). 오버행 영역의 높은 표면 거칠기는 (마이크로) 정확도 / 품질에 대한 엄격한 요구 사항이있는 부품 제조에서 L-PBF의 적용을 제한합니다.

본 연구는 실험 및 시뮬레이션 연구를 사용하여 오버행 영역 (지지물없이 제작)의 미세 형상 편차 형성 메커니즘과 표면 거칠기의 기원을 체계적이고 포괄적으로 조사합니다. 결합 된 DEM-CFD 시뮬레이션 모델은 경계 트랙 윤곽, 분말 접착 및 뒤틀림 변형의 효과를 고려하여 오버행 영역의 용융 풀 역학과 표면 형태의 형성 메커니즘을 나타 내기 위해 개발되었습니다. 표면 거칠기 R의 시뮬레이션 및 단일 요인 L-PBF 인쇄 실험을 사용하여 오버행 각도의 함수로 연구됩니다. 용융 풀의 침몰과 관련된 오버행 영역에서 분말 접착의 세 가지 메커니즘이 식별되고 자세히 설명됩니다. 마지막으로, 인쇄 된 오버행 영역에서 높은 표면 거칠기 문제를 완화 할 수 있는 잠재적 솔루션에 대해 간략하게 설명합니다.

The shape and size of the L-PBF printed samples are illustrated in Figure 1
The shape and size of the L-PBF printed samples are illustrated in Figure 1
Figure 2. Borders in the overhang region depending on the overhang angle θ
Figure 2. Borders in the overhang region depending on the overhang angle θ
Figure 3. (a) Profile of the volumetric heat source, (b) the model geometry of single-track printing on a solid substrate (unit: µm), and (c) the comparison of melt pool dimensions obtained from the experiment (right half) and simulation (left half) for a calibrated optical penetration depth of 110 µm (laser power 200 W and scan speed 800 mm/s, solidified layer thickness 30 µm, powder size 10–45 µm).
Figure 3. (a) Profile of the volumetric heat source, (b) the model geometry of single-track printing on a solid substrate (unit: µm), and (c) the comparison of melt pool dimensions obtained from the experiment (right half) and simulation (left half) for a calibrated optical penetration depth of 110 µm (laser power 200 W and scan speed 800 mm/s, solidified layer thickness 30 µm, powder size 10–45 µm).
Figure 4. The model geometry of an overhang being L-PBF processed: (a) 3D view and (b) right view.
Figure 4. The model geometry of an overhang being L-PBF processed: (a) 3D view and (b) right view.
Figure 5. The cross-sectional contour of border tracks in a 45° overhang region.
Figure 5. The cross-sectional contour of border tracks in a 45° overhang region.
Figure 6. Evolution of melt pool in the overhang region (θ = 45°, P = 100 W, v = 1000 mm/s, the streamlines are shown by arrows).
Figure 6. Evolution of melt pool in the overhang region (θ = 45°, P = 100 W, v = 1000 mm/s, the streamlines are shown by arrows).
Figure 7. The overhang contour is contributed by (a) only outer borders when θ ≤ 60° (b) both inner borders and outer borders when θ > 60°.
Figure 7. The overhang contour is contributed by (a) only outer borders when θ ≤ 60° (b) both inner borders and outer borders when θ > 60°.
Figure 8. Schematic of powder adhesion on a 45° overhang region.
Figure 8. Schematic of powder adhesion on a 45° overhang region.
Figure 9. The L-PBF printed samples with various overhang angle (a) θ = 0° (cube), (b) θ = 30°, (c) θ = 45°, (d) θ = 55° and (e) θ = 60°.
Figure 9. The L-PBF printed samples with various overhang angle (a) θ = 0° (cube), (b) θ = 30°, (c) θ = 45°, (d) θ = 55° and (e) θ = 60°.
Figure 10. Two mechanisms of powder adhesion related to the overhang angle: (a) simulation-predicted, θ = 45°; (b) simulation-predicted, θ = 60°; (c, e) optical micrographs, θ = 45°; (d, f) optical micrographs, θ = 60°. (e) and (f) are partial enlargement of (c) and (d), respectively.
Figure 10. Two mechanisms of powder adhesion related to the overhang angle: (a) simulation-predicted, θ = 45°; (b) simulation-predicted, θ = 60°; (c, e) optical micrographs, θ = 45°; (d, f) optical micrographs, θ = 60°. (e) and (f) are partial enlargement of (c) and (d), respectively.
Figure 11. Simulation-predicted surface morphology in the overhang region at different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45°, (d) θ = 60° and (e) θ = 80° (Blue solid lines: simulation-predicted contour; red dashed lines: the planar profile of designed overhang region specified by the overhang angles).
Figure 11. Simulation-predicted surface morphology in the overhang region at different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45°, (d) θ = 60° and (e) θ = 80° (Blue solid lines: simulation-predicted contour; red dashed lines: the planar profile of designed overhang region specified by the overhang angles).
Figure 12. Effect of overhang angle on surface roughness Ra in overhang regions
Figure 12. Effect of overhang angle on surface roughness Ra in overhang regions
Figure 13. Surface morphology of L-PBF printed overhang regions with different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45° and (d) θ = 60° (overhang border parameters: P = 100 W, v = 1000 mm/s).
Figure 13. Surface morphology of L-PBF printed overhang regions with different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45° and (d) θ = 60° (overhang border parameters: P = 100 W, v = 1000 mm/s).
Figure 14. Effect of (a) laser power (scan speed = 1000 mm/s) and (b) scan speed (lase power = 100 W) on surface roughness Ra in overhang regions (θ = 45°, laser power and scan speed referred to overhang border parameters, and the other process parameters are listed in Table 2).
Figure 14. Effect of (a) laser power (scan speed = 1000 mm/s) and (b) scan speed (lase power = 100 W) on surface roughness Ra in overhang regions (θ = 45°, laser power and scan speed referred to overhang border parameters, and the other process parameters are listed in Table 2).

References

  • Cai, Chao, Chrupcala Radoslaw, Jinliang Zhang, Qian Yan, Shifeng Wen, Bo Song, and Yusheng Shi. 2019. “In-Situ Preparation and Formation of TiB/Ti-6Al-4V Nanocomposite via Laser Additive Manufacturing: Microstructure Evolution and Tribological Behavior.” Powder Technology 342: 73–84. doi:10.1016/j.powtec.2018.09.088. [Crossref], [Web of Science ®], [Google Scholar]
  • Cai, Chao, Wei Shian Tey, Jiayao Chen, Wei Zhu, Xingjian Liu, Tong Liu, Lihua Zhao, and Kun Zhou. 2021. “Comparative Study on 3D Printing of Polyamide 12 by Selective Laser Sintering and Multi Jet Fusion.” Journal of Materials Processing Technology 288 (August 2020): 116882. doi:10.1016/j.jmatprotec.2020.116882. [Crossref], [Web of Science ®], [Google Scholar]
  • Cai, Chao, Xu Wu, Wan Liu, Wei Zhu, Hui Chen, Jasper Dong Qiu Chua, Chen Nan Sun, Jie Liu, Qingsong Wei, and Yusheng Shi. 2020. “Selective Laser Melting of Near-α Titanium Alloy Ti-6Al-2Zr-1Mo-1V: Parameter Optimization, Heat Treatment and Mechanical Performance.” Journal of Materials Science and Technology 57: 51–64. doi:10.1016/j.jmst.2020.05.004. [Crossref], [Web of Science ®], [Google Scholar]
  • Charles, Amal, Ahmed Elkaseer, Lore Thijs, and Steffen G. Scholz. 2020. “Dimensional Errors Due to Overhanging Features in Laser Powder Bed Fusion Parts Made of Ti-6Al-4V.” Applied Sciences 10 (7): 2416. doi:10.3390/app10072416. [Crossref], [Google Scholar]
  • Chatham, Camden A., Timothy E. Long, and Christopher B. Williams. 2019. “A Review of the Process Physics and Material Screening Methods for Polymer Powder Bed Fusion Additive Manufacturing.” Progress in Polymer Science 93: 68–95. doi:10.1016/j.progpolymsci.2019.03.003. [Crossref], [Web of Science ®], [Google Scholar]
  • Du, Yang, Xinyu You, Fengbin Qiao, Lijie Guo, and Zhengwu Liu. 2019. “A Model for Predicting the Temperature Field during Selective Laser Melting.” Results in Physics 12 (November 2018): 52–60. doi:10.1016/j.rinp.2018.11.031. [Crossref], [Web of Science ®], [Google Scholar]
  • Duval-Chaneac, M. S., S. Han, C. Claudin, F. Salvatore, J. Bajolet, and J. Rech. 2018. “Experimental Study on Finishing of Internal Laser Melting (SLM) Surface with Abrasive Flow Machining (AFM).” Precision Engineering 54 (July 2017): 1–6. doi:10.1016/j.precisioneng.2018.03.006. [Crossref], [Web of Science ®], [Google Scholar]
  • Feng, Shaochuan, Shijie Chen, Amar M. Kamat, Ru Zhang, Mingji Huang, and Liangcai Hu. 2020. “Investigation on Shape Deviation of Horizontal Interior Circular Channels Fabricated by Laser Powder Bed Fusion.” Additive Manufacturing 36 (December): 101585. doi:10.1016/j.addma.2020.101585. [Crossref], [Web of Science ®], [Google Scholar]
  • Feng, Shaochuan, Chuanzhen Huang, Jun Wang, Hongtao Zhu, Peng Yao, and Zhanqiang Liu. 2017. “An Analytical Model for the Prediction of Temperature Distribution and Evolution in Hybrid Laser-Waterjet Micro-Machining.” Precision Engineering 47: 33–45. doi:10.1016/j.precisioneng.2016.07.002. [Crossref], [Web of Science ®], [Google Scholar]
  • Feng, Shaochuan, Amar M. Kamat, and Yutao Pei. 2021. “Design and Fabrication of Conformal Cooling Channels in Molds: Review and Progress Updates.” International Journal of Heat and Mass Transfer. doi:10.1016/j.ijheatmasstransfer.2021.121082. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
  • Flow-3D V11.2 Documentation. 2016. Flow Science, Inc. [Crossref], [Google Scholar]
  • Foroozmehr, Ali, Mohsen Badrossamay, Ehsan Foroozmehr, and Sa’id Golabi. 2016. “Finite Element Simulation of Selective Laser Melting Process Considering Optical Penetration Depth of Laser in Powder Bed.” Materials and Design 89: 255–263. doi:10.1016/j.matdes.2015.10.002. [Crossref], [Web of Science ®], [Google Scholar]
  • “Geometrical Product Specifications (GPS) — Surface Texture: Profile Method — Rules and Procedures for the Assessment of Surface Texture (ISO 4288).” 1996. International Organization for Standardization. https://www.iso.org/standard/2096.html. [Google Scholar]
  • Günther, Johannes, Stefan Leuders, Peter Koppa, Thomas Tröster, Sebastian Henkel, Horst Biermann, and Thomas Niendorf. 2018. “On the Effect of Internal Channels and Surface Roughness on the High-Cycle Fatigue Performance of Ti-6Al-4V Processed by SLM.” Materials & Design 143: 1–11. doi:10.1016/j.matdes.2018.01.042. [Crossref], [Web of Science ®], [Google Scholar]
  • Hopkinson, Neil, and Phill Dickens. 2000. “Conformal Cooling and Heating Channels Using Laser Sintered Tools.” In Solid Freeform Fabrication Conference, 490–497. Texas. doi:10.26153/tsw/3075. [Crossref], [Google Scholar]
  • Hu, Zhiheng, Haihong Zhu, Changchun Zhang, Hu Zhang, Ting Qi, and Xiaoyan Zeng. 2018. “Contact Angle Evolution during Selective Laser Melting.” Materials and Design 139: 304–313. doi:10.1016/j.matdes.2017.11.002. [Crossref], [Web of Science ®], [Google Scholar]
  • Hu, Cheng, Kejia Zhuang, Jian Weng, and Donglin Pu. 2019. “Three-Dimensional Analytical Modeling of Cutting Temperature for Round Insert Considering Semi-Infinite Boundary and Non-Uniform Heat Partition.” International Journal of Mechanical Sciences 155 (October 2018): 536–553. doi:10.1016/j.ijmecsci.2019.03.019. [Crossref], [Web of Science ®], [Google Scholar]
  • Huang, Yuze, Mir Behrad Khamesee, and Ehsan Toyserkani. 2019. “A New Physics-Based Model for Laser Directed Energy Deposition (Powder-Fed Additive Manufacturing): From Single-Track to Multi-Track and Multi-Layer.” Optics & Laser Technology 109 (August 2018): 584–599. doi:10.1016/j.optlastec.2018.08.015. [Crossref], [Web of Science ®], [Google Scholar]
  • Kadirgama, K., W. S. W. Harun, F. Tarlochan, M. Samykano, D. Ramasamy, Mohd Zaidi Azir, and H. Mehboob. 2018. “Statistical and Optimize of Lattice Structures with Selective Laser Melting (SLM) of Ti6AL4V Material.” International Journal of Advanced Manufacturing Technology 97 (1–4): 495–510. doi:10.1007/s00170-018-1913-1. [Crossref], [Web of Science ®], [Google Scholar]
  • Kamat, Amar M, and Yutao Pei. 2019. “An Analytical Method to Predict and Compensate for Residual Stress-Induced Deformation in Overhanging Regions of Internal Channels Fabricated Using Powder Bed Fusion.” Additive Manufacturing 29 (March): 100796. doi:10.1016/j.addma.2019.100796. [Crossref], [Web of Science ®], [Google Scholar]
  • Kamath, Chandrika, Bassem El-Dasher, Gilbert F. Gallegos, Wayne E. King, and Aaron Sisto. 2014. “Density of Additively-Manufactured, 316L SS Parts Using Laser Powder-Bed Fusion at Powers up to 400 W.” International Journal of Advanced Manufacturing Technology 74 (1–4): 65–78. doi:10.1007/s00170-014-5954-9. [Crossref], [Web of Science ®], [Google Scholar]
  • Karimi, J., C. Suryanarayana, I. Okulov, and K. G. Prashanth. 2020. “Selective Laser Melting of Ti6Al4V: Effect of Laser Re-Melting.” Materials Science and Engineering A (July): 140558. doi:10.1016/j.msea.2020.140558. [Crossref], [Web of Science ®], [Google Scholar]
  • Khairallah, Saad A., and Andy Anderson. 2014. “Mesoscopic Simulation Model of Selective Laser Melting of Stainless Steel Powder.” Journal of Materials Processing Technology 214 (11): 2627–2636. doi:10.1016/j.jmatprotec.2014.06.001. [Crossref], [Web of Science ®], [Google Scholar]
  • Khairallah, Saad A., Andrew T. Anderson, Alexander Rubenchik, and Wayne E. King. 2016. “Laser Powder-Bed Fusion Additive Manufacturing: Physics of Complex Melt Flow and Formation Mechanisms of Pores, Spatter, and Denudation Zones.” Edited by Adedeji B. Badiru, Vhance V. Valencia, and David Liu. Acta Materialia 108 (April): 36–45. doi:10.1016/j.actamat.2016.02.014. [Crossref], [Web of Science ®], [Google Scholar]
  • Kuo, C. N., C. K. Chua, P. C. Peng, Y. W. Chen, S. L. Sing, S. Huang, and Y. L. Su. 2020. “Microstructure Evolution and Mechanical Property Response via 3D Printing Parameter Development of Al–Sc Alloy.” Virtual and Physical Prototyping 15 (1): 120–129. doi:10.1080/17452759.2019.1698967. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Le, K. Q., C. H. Wong, K. H. G. Chua, C. Tang, and H. Du. 2020. “Discontinuity of Overhanging Melt Track in Selective Laser Melting Process.” International Journal of Heat and Mass Transfer 162 (December): 120284. doi:10.1016/j.ijheatmasstransfer.2020.120284. [Crossref], [Web of Science ®], [Google Scholar]
  • Lee, Y. S., and W. Zhang. 2016. “Modeling of Heat Transfer, Fluid Flow and Solidification Microstructure of Nickel-Base Superalloy Fabricated by Laser Powder Bed Fusion.” Additive Manufacturing 12: 178–188. doi:10.1016/j.addma.2016.05.003. [Crossref], [Web of Science ®], [Google Scholar]
  • Leitz, K. H., P. Singer, A. Plankensteiner, B. Tabernig, H. Kestler, and L. S. Sigl. 2017. “Multi-Physical Simulation of Selective Laser Melting.” Metal Powder Report 72 (5): 331–338. doi:10.1016/j.mprp.2016.04.004. [Crossref], [Google Scholar]
  • Li, Jian, Jing Hu, Yi Zhu, Xiaowen Yu, Mengfei Yu, and Huayong Yang. 2020. “Surface Roughness Control of Root Analogue Dental Implants Fabricated Using Selective Laser Melting.” Additive Manufacturing 34 (September 2019): 101283. doi:10.1016/j.addma.2020.101283. [Crossref], [Web of Science ®], [Google Scholar]
  • Li, Yingli, Kun Zhou, Pengfei Tan, Shu Beng Tor, Chee Kai Chua, and Kah Fai Leong. 2018. “Modeling Temperature and Residual Stress Fields in Selective Laser Melting.” International Journal of Mechanical Sciences 136 (February): 24–35. doi:10.1016/j.ijmecsci.2017.12.001. [Crossref], [Web of Science ®], [Google Scholar]
  • Mazur, MacIej, Martin Leary, Matthew McMillan, Joe Elambasseril, and Milan Brandt. 2016. “SLM Additive Manufacture of H13 Tool Steel with Conformal Cooling and Structural Lattices.” Rapid Prototyping Journal 22 (3): 504–518. doi:10.1108/RPJ-06-2014-0075. [Crossref], [Web of Science ®], [Google Scholar]
  • Mingear, Jacob, Bing Zhang, Darren Hartl, and Alaa Elwany. 2019. “Effect of Process Parameters and Electropolishing on the Surface Roughness of Interior Channels in Additively Manufactured Nickel-Titanium Shape Memory Alloy Actuators.” Additive Manufacturing 27 (October 2018): 565–575. doi:10.1016/j.addma.2019.03.027. [Crossref], [Web of Science ®], [Google Scholar]
  • Pakkanen, Jukka, Flaviana Calignano, Francesco Trevisan, Massimo Lorusso, Elisa Paola Ambrosio, Diego Manfredi, and Paolo Fino. 2016. “Study of Internal Channel Surface Roughnesses Manufactured by Selective Laser Melting in Aluminum and Titanium Alloys.” Metallurgical and Materials Transactions A 47 (8): 3837–3844. doi:10.1007/s11661-016-3478-7. [Crossref], [Web of Science ®], [Google Scholar]
  • Patterson, Albert E., Sherri L. Messimer, and Phillip A. Farrington. 2017. “Overhanging Features and the SLM/DMLS Residual Stresses Problem: Review and Future Research Need.” Technologies 5 (4): 15. doi:10.3390/technologies5020015. [Crossref], [Web of Science ®], [Google Scholar]
  • Roberts, I. A., C. J. Wang, R. Esterlein, M. Stanford, and D. J. Mynors. 2009. “A Three-Dimensional Finite Element Analysis of the Temperature Field during Laser Melting of Metal Powders in Additive Layer Manufacturing.” International Journal of Machine Tools and Manufacture 49 (12–13): 916–923. doi:10.1016/j.ijmachtools.2009.07.004. [Crossref], [Web of Science ®], [Google Scholar]
  • Shrestha, Subin, and Kevin Chou. 2018. “Computational Analysis of Thermo-Fluid Dynamics with Metallic Powder in SLM.” In CFD Modeling and Simulation in Materials Processing 2018, edited by Laurentiu Nastac, Koulis Pericleous, Adrian S. Sabau, Lifeng Zhang, and Brian G. Thomas, 85–95. Cham, Switzerland: Springer Nature. doi:10.1007/978-3-319-72059-3_9. [Crossref], [Google Scholar]
  • Sing, S. L., and W. Y. Yeong. 2020. “Laser Powder Bed Fusion for Metal Additive Manufacturing: Perspectives on Recent Developments.” Virtual and Physical Prototyping 15 (3): 359–370. doi:10.1080/17452759.2020.1779999. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
  • Šmilauer, Václav, Emanuele Catalano, Bruno Chareyre, Sergei Dorofeenko, Jérôme Duriez, Nolan Dyck, Jan Eliáš, et al. 2015. Yade Documentation. 2nd ed. The Yade Project. doi:10.5281/zenodo.34073. [Crossref], [Google Scholar]
  • Tan, Pengfei, Fei Shen, Biao Li, and Kun Zhou. 2019. “A Thermo-Metallurgical-Mechanical Model for Selective Laser Melting of Ti6Al4V.” Materials & Design 168 (April): 107642. doi:10.1016/j.matdes.2019.107642. [Crossref], [Web of Science ®], [Google Scholar]
  • Tan, Lisa Jiaying, Wei Zhu, and Kun Zhou. 2020. “Recent Progress on Polymer Materials for Additive Manufacturing.” Advanced Functional Materials 30 (43): 1–54. doi:10.1002/adfm.202003062. [Crossref], [Web of Science ®], [Google Scholar]
  • Wang, Xiaoqing, and Kevin Chou. 2018. “Effect of Support Structures on Ti-6Al-4V Overhang Parts Fabricated by Powder Bed Fusion Electron Beam Additive Manufacturing.” Journal of Materials Processing Technology 257 (February): 65–78. doi:10.1016/j.jmatprotec.2018.02.038. [Crossref], [Web of Science ®], [Google Scholar]
  • Wang, Di, Yongqiang Yang, Ziheng Yi, and Xubin Su. 2013. “Research on the Fabricating Quality Optimization of the Overhanging Surface in SLM Process.” International Journal of Advanced Manufacturing Technology 65 (9–12): 1471–1484. doi:10.1007/s00170-012-4271-4. [Crossref], [Web of Science ®], [Google Scholar]
  • Wen, Peng, Maximilian Voshage, Lucas Jauer, Yanzhe Chen, Yu Qin, Reinhart Poprawe, and Johannes Henrich Schleifenbaum. 2018. “Laser Additive Manufacturing of Zn Metal Parts for Biodegradable Applications: Processing, Formation Quality and Mechanical Properties.” Materials and Design 155: 36–45. doi:10.1016/j.matdes.2018.05.057. [Crossref], [Web of Science ®], [Google Scholar]
  • Wu, Yu-che, Cheng-hung San, Chih-hsiang Chang, Huey-jiuan Lin, Raed Marwan, Shuhei Baba, and Weng-Sing Hwang. 2018. “Numerical Modeling of Melt-Pool Behavior in Selective Laser Melting with Random Powder Distribution and Experimental Validation.” Journal of Materials Processing Technology 254 (November 2017): 72–78. doi:10.1016/j.jmatprotec.2017.11.032. [Crossref], [Web of Science ®], [Google Scholar]
  • Yadroitsev, I., P. Krakhmalev, I. Yadroitsava, S. Johansson, and I. Smurov. 2013. “Energy Input Effect on Morphology and Microstructure of Selective Laser Melting Single Track from Metallic Powder.” Journal of Materials Processing Technology 213 (4): 606–613. doi:10.1016/j.jmatprotec.2012.11.014. [Crossref], [Web of Science ®], [Google Scholar]
  • Yu, Wenhui, Swee Leong Sing, Chee Kai Chua, and Xuelei Tian. 2019. “Influence of Re-Melting on Surface Roughness and Porosity of AlSi10Mg Parts Fabricated by Selective Laser Melting.” Journal of Alloys and Compounds 792: 574–581. doi:10.1016/j.jallcom.2019.04.017. [Crossref], [Web of Science ®], [Google Scholar]
  • Zhang, Dongyun, Pudan Zhang, Zhen Liu, Zhe Feng, Chengjie Wang, and Yanwu Guo. 2018. “Thermofluid Field of Molten Pool and Its Effects during Selective Laser Melting (SLM) of Inconel 718 Alloy.” Additive Manufacturing 21 (100): 567–578. doi:10.1016/j.addma.2018.03.031. [Crossref], [Web of Science ®], [Google Scholar]
Dam-Break Flows: Comparison between Flow-3D, MIKE 3 FM, and Analytical Solutions with Experimental Data

Dam-Break Flows: Comparison between Flow-3D, MIKE 3 FM, and Analytical Solutions with Experimental Data

by Hui Hu,Jianfeng Zhang andTao Li *
State Key Laboratory Base of Eco-Hydraulic Engineering in Arid Area, School of Water Resources and Hydropower, Xi’an University of Technology, Xi’an 710048, China
*Author to whom correspondence should be addressed.
Appl. Sci.20188(12), 2456; https://doi.org/10.3390/app8122456Received: 14 October 2018 /
Revised: 20 November 2018 / Accepted: 29 November 2018 / Published: 2 December 2018

Abstract

The objective of this study was to evaluate the applicability of a flow model with different numbers of spatial dimensions in a hydraulic features solution, with parameters such a free surface profile, water depth variations, and averaged velocity evolution in a dam-break under dry and wet bed conditions with different tailwater depths. Two similar three-dimensional (3D) hydrodynamic models (Flow-3D and MIKE 3 FM) were studied in a dam-break simulation by performing a comparison with published experimental data and the one-dimensional (1D) analytical solution. The results indicate that the Flow-3D model better captures the free surface profile of wavefronts for dry and wet beds than other methods. The MIKE 3 FM model also replicated the free surface profiles well, but it underestimated them during the initial stage under wet-bed conditions. However, it provided a better approach to the measurements over time. Measured and simulated water depth variations and velocity variations demonstrate that both of the 3D models predict the dam-break flow with a reasonable estimation and a root mean square error (RMSE) lower than 0.04, while the MIKE 3 FM had a small memory footprint and the computational time of this model was 24 times faster than that of the Flow-3D. Therefore, the MIKE 3 FM model is recommended for computations involving real-life dam-break problems in large domains, leaving the Flow-3D model for fine calculations in which knowledge of the 3D flow structure is required. The 1D analytical solution was only effective for the dam-break wave propagations along the initially dry bed, and its applicability was fairly limited. 

Keywords: dam breakFlow-3DMIKE 3 FM1D Ritter’s analytical solution

이 연구의 목적은 자유 표면 프로파일, 수심 변화 및 건식 및 댐 파괴에서 평균 속도 변화와 같은 매개 변수를 사용하여 유압 기능 솔루션에서 서로 다른 수의 공간 치수를 가진 유동 모델의 적용 가능성을 평가하는 것이었습니다.

테일 워터 깊이가 다른 습식베드 조건. 2 개의 유사한 3 차원 (3D) 유체 역학 모델 (Flow-3D 및 MIKE 3 FM)이 게시된 실험 데이터와 1 차원 (1D) 분석 솔루션과의 비교를 수행하여 댐 브레이크 시뮬레이션에서 연구되었습니다.

결과는 FLOW-3D 모델이 다른 방법보다 건식 및 습식 베드에 대한 파면의 자유 표면 프로파일을 더 잘 포착함을 나타냅니다. MIKE 3 FM 모델도 자유 표면 프로파일을 잘 복제했지만, 습식 조건에서 초기 단계에서 과소 평가했습니다. 그러나 시간이 지남에 따라 측정에 더 나은 접근 방식을 제공했습니다.

측정 및 시뮬레이션 된 수심 변화와 속도 변화는 두 3D 모델 모두 합리적인 추정치와 0.04보다 낮은 RMSE (root mean square error)로 댐 브레이크 흐름을 예측하는 반면 MIKE 3 FM은 메모리 공간이 적고 이 모델의 계산 시간은 Flow-3D보다 24 배 더 빠릅니다.

따라서 MIKE 3 FM 모델은 대규모 도메인의 실제 댐 브레이크 문제와 관련된 계산에 권장되며 3D 흐름 구조에 대한 지식이 필요한 미세 계산을 위해 Flow-3D 모델을 남겨 둡니다. 1D 분석 솔루션은 초기 건조 층을 따라 전파되는 댐 파괴에만 효과적이었으며 그 적용 가능성은 상당히 제한적이었습니다.

1. Introduction

저수지에 저장된 물의 통제되지 않은 방류[1]로 인해 댐 붕괴와 그로 인해 하류에서 발생할 수 있는 잠재적 홍수로 인해 큰 자연 위험이 발생한다. 이러한 영향을 최대한 완화하기 위해서는 홍수[2]로 인한 위험을 관리하고 감소시키기 위해 홍수의 시간적 및 공간적 진화를 모두 포착하여 댐 붕괴 파동의 움직임을 예측하고 댐 붕괴 파동의 전파 과정 효과를 다운스트림[3]으로 예측하는 것이 중요하다. 

그러나 이러한 수량을 예측하는 것은 어려운 일이며, 댐 붕괴 홍수의 움직임을 정확하게 시뮬레이션하고 유동장에 대한 유용한 정보를 제공하기 위한 적절한 모델을 선택하는 것은 그러므로 필수적인 단계[4]이다.

적절한 수학적 및 수치적 모델의 선택은 댐 붕괴 홍수 분석에서 매우 중요한 것으로 나타났다.분석적 해결책에서 행해진 댐 붕괴 흐름에 대한 연구는 100여 년 전에 시작되었다. 

리터[5]는 먼저 건조한 침대 위에 1D de 생베넌트 방정식의 초기 분석 솔루션을 도출했고, 드레슬러[6,7]와 휘담[8]은 마찰저항의 영향을 받은 파동학을 연구했으며, 스토커[9]는 젖은 침대를 위한 1D 댐 붕괴 문제에 리터의 솔루션을 확장했다. 

마샬과 멩데즈[10]는 고두노프가 가스 역학의 오일러 방정식을 위해 개발한 방법론[11]을 적용하여 젖은 침대 조건에서 리만 문제를 해결하기 위한 일반적인 절차를 고안했다. Toro [12]는 습식 및 건식 침대 조건을 모두 해결하기 위해 완전한 1D 정밀 리만 용해제를 실시했다. 

Chanson [13]은 특성 방법을 사용하여 갑작스러운 댐 붕괴로 인한 홍수에 대한 간단한 분석 솔루션을 연구했다. 그러나 이러한 분석 솔루션은 특히 댐 붕괴 초기 단계에서 젖은 침대의 정확한 결과를 도출하지 못했다[14,15].과거 연구의 발전은 이른바 댐 붕괴 홍수 문제 해결을 위한 여러 수치 모델[16]을 제공했으며, 헥-라스, DAMBRK, MIK 11 등과 같은 1차원 모델을 댐 붕괴 홍수를 모델링하는 데 사용하였다.

[17 2차원(2D) 깊이 평균 방정식도 댐 붕괴 흐름 문제를 시뮬레이션하는 데 널리 사용되어 왔으며[18,19,20,21,22] 그 결과 얕은 물 방정식(SWE)이 유체 흐름을 나타내는 데 적합하다는 것을 알 수 있다. 그러나, 경우에 따라 2D 수치해결기가 제공하는 해결책이 특히 근거리 분야에서 실험과 일관되지 않을 수 있다[23,24]. 더욱이, 1차원 및 2차원 모델은 3차원 현상에 대한 일부 세부사항을 포착하는 데 한계가 있다.

[25]. RANS(Reynolds-averageed Navier-Stok크스 방정식)에 기초한 여러 3차원(3D) 모델이 얕은 물 모델의 일부 단점을 극복하기 위해 적용되었으며, 댐 붕괴 초기 단계에서의 복잡한 흐름의 실제 동작을 이해하기 위해 사용되었다 [26,27,28]장애물이나 바닥 실에 대한 파장의 충격으로 인한 튜디 댐 붕괴 흐름 [19,29] 및 근거리 영역의 난류 댐 붕괴 흐름 거동 [4] 최근 상용화된 수치 모델 중 잘 알려진 유체 방식(VOF) 기반 CFD 모델링 소프트웨어 FLOW-3D는 컴퓨터 기술의 진보에 따른 계산력 증가로 인해 불안정한 자유 표면 흐름을 분석하는 데 널리 사용되고 있다. 

이 소프트웨어는 유한 차이 근사치를 사용하여 RANS 방정식에 대한 수치 해결책을 계산하며, 자유 표면을 추적하기 위해 VOF를 사용한다 [30,31]; 댐 붕괴 흐름을 모델링하는 데 성공적으로 사용되었다 [32,33].그러나, 2D 얕은 물 모델을 사용하여 포착할 수 없는 공간과 시간에 걸친 댐 붕괴 흐름의 특정한 유압적 특성이 있다. 

실생활 현장 척도 시뮬레이션을 위한 완전한 3D Navier-Stokes 방정식의 적용은 더 높은 계산 비용[34]을 가지고 있으며, 원하는 결과는 얕은 물 모델[35]보다 더 정확한 결과를 산출하지 못할 수 있다. 따라서, 본 논문은 3D 모델의 기능과 그 계산 효율을 평가하기 위해 댐 붕괴 흐름 시뮬레이션을 위한 단순화된 3D 모델-MIKE 3 FM을 시도한다. 

MIK 3 모델은 자연 용수 분지의 여러 유체 역학 시뮬레이션 조사에 적용되었다. 보치 외 연구진이 사용해 왔다. [36], 니콜라오스 및 게오르기오스 [37], 고얄과 라토드[38] 등 현장 연구에서 유체역학 시뮬레이션을 위한 것이다. 이러한 저자들의 상당한 연구에도 불구하고, MIK 3 FM을 이용한 댐 붕괴의 모델링에 관한 연구는 거의 없었다. 

또한 댐 붕괴 홍수 전파 문제를 해결하기 위한 3D 얕은 물과 완전한 3D RANS 모델의 성능을 비교한 연구도 아직 보고되지 않았다. 이 공백을 메우기 위해 현재 연구의 주요 목표는 댐 붕괴 흐름을 시뮬레이션하기 위한 단순화된 3D SWE, 상세 RANS 모델 및 분석 솔루션을 평가하여 댐 붕괴 문제에 대한 정확도와 적용 가능성을 평가하는 것이다.실제 댐 붕괴 문제를 해결하기 위해 유체역학 시뮬레이션을 시도하기 전에 수치 모델을 검증할 필요가 있다. 

일련의 실험 벤치마크를 사용하여 수치 모델을 확인하는 것은 용인된 관행이다. 현장 데이터 확보가 어려워 최근 몇 년 동안 제한된 측정 데이터를 취득했다. 

본 논문은 Ozmen-Cagatay와 Kocaman[30] 및 Khankandi 외 연구진이 제안한 두 가지 테스트 사례에 의해 제안된 검증에서 인용한 것이다. [39] 오즈멘-카가테이와 코카만[30]이 수행한 첫 번째 실험에서, 다른 미숫물 수위에 걸쳐 초기 단계 동안 댐 붕괴 홍수파가 발생했으며, 자유 지표면 프로파일의 측정치를 제공했다. Ozmen-Cagatay와 Kocaman[30]은 초기 단계에서 Flow-3D 소프트웨어가 포함된 2D SWE와 3D RANS의 숫자 솔루션에 의해 계산된 자유 표면 프로필만 비교했다. 

Khankandi 등이 고안한 두 번째 실험 동안. [39], 이 실험의 측정은 홍수 전파를 시뮬레이션하고 측정된 데이터를 제공하는 것을 목적으로 하는 수치 모델을 검증하기 위해 사용되었으며, 말기 동안의 자유 표면 프로필, 수위의 시간 진화 및 속도 변화를 포함한다. Khankandi 등의 연구. [39] 주로 실험 조사에 초점을 맞추었으며, 초기 단계에서는 리터의 솔루션과의 수위만을 언급하고 있다.

경계 조건(상류 및 하류 모두 무한 채널 길이를 갖는 1D 분석 솔루션에서는 실험 결과를 리터와 비교하는 것이 타당하지 않기 때문이다(건조 be)d) 또는 스토커(웨트 베드) 솔루션은 벽의 반사가 깊이 프로파일에 영향을 미쳤을 때, 그리고 참조 [39]의 실험에 대한 수치 시뮬레이션과의 추가 비교가 불량할 때. 이 논문은 이러한 문제를 직접 겨냥하여 전체 댐 붕괴 과정에서의 자유 표면 프로필, 수심 변화 및 속도 변화에 대한 완전한 비교 연구를 제시한다. 

여기서 댐 붕괴파의 수치 시뮬레이션은 초기에 건조하고 습한 직사각형 채널을 가진 유한 저장소의 순간 댐 붕괴에 대해 두 개의 3D 모델을 사용하여 개발된다.본 논문은 다음과 같이 정리되어 있다. 두 모델에 대한 통치 방정식은 숫자 체계를 설명하기 전에 먼저 도입된다. 

일반적인 단순화된 시험 사례는 3D 수치 모델과 1D 분석 솔루션을 사용하여 시뮬레이션했다. 모델 결과와 이들이 실험실 실험과 비교하는 방법이 논의되고, 서로 다른 수심비에서 시간에 따른 유압 요소의 변동에 대한 시뮬레이션 결과가 결론을 도출하기 전에 제시된다.

2. Materials and Methods

2.1. Data

첫째, 수평 건조 및 습식 침상에 대한 초기 댐 붕괴 단계 동안의 자유 표면 프로필 측정은 Ozmen-Cagatay와 Kocaman에 의해 수행되었다[30]. 이 시험 동안, 매끄럽고 직사각형의 수평 채널은 그림 1에서 표시한 대로 너비 0.30m, 높이 0.30m, 길이 8.9m이었다. 

채널은 채널 입구에서 4.65m 떨어진 수직 플레이트(담) 즉, 저장소의 길이 L0=4.65mL0에 의해 분리되었다., 및 다운스트림 채널 L1=4.25 mL1. m저수지는 댐의 좌측에 위치하고 처음에는 침수된 것으로 간주되었다; 저수지의 초기 상류 수심 h0 0.25m로 일정했다.

오른쪽의 초기 수심 h1h1 건식침대의 경우 0m, 습식침대의 경우 0.025m, 0.1m이므로 수심비 α=h1/h0α으로 세 가지 상황이 있었다. 0, 0.1, 0.4의 습식침대 조건은 플룸 끝에 낮은 보를 사용함으로써 만들어졌다. 물 표면 프로필은 3개의 고속 디지털 카메라(50프레임/s)를 사용하여 초기에 관찰되었으며, 계측 측정의 정확도는 참고문헌 [30]에서 입증되었다. In the following section, the corresponding numerical results refer to positions x = −1 m (P1), −0.5 m (P2), −0.2 m (P3), +0.2 m (P4), +0.5 m (P5), +1 m (P6), +2 m (P7), and +2.85 m (P8), where the origin of the coordinate system x = 0 is at the dam site. 3수심비 ααα 0, 0.1, 0.4의 경우 x,yx의 경우 좌표는 h0.으로 정규화된다.

<중략> ……

Figure 1. Schematic view of the experimental conditions by Ozmen-Cagatay and Kocaman [30]: (a) α = 0; (b) α = 0.1; and (c) α = 0.4.
Figure 1. Schematic view of the experimental conditions by Ozmen-Cagatay and Kocaman [30]: (a) α = 0; (b) α = 0.1; and (c) α = 0.4.

Figure 2. Schematic view of the experimental conditions by Khankandi et al. [39]: (a) α = 0 and (b) α = 0.2.
Figure 2. Schematic view of the experimental conditions by Khankandi et al. [39]: (a) α = 0 and (b) α = 0.2.
Figure 3. Typical profiles of the dam-break flow regimes for Stoker’s analytical solution [9]: Wet-bed downstream
Figure 3. Typical profiles of the dam-break flow regimes for Stoker’s analytical solution [9]: Wet-bed downstream
Figure 4. Sensitivity analysis of the numerical simulation using Flow-3D for the different mesh sizes of the experiments in Reference [30].
Figure 4. Sensitivity analysis of the numerical simulation using Flow-3D for the different mesh sizes of the experiments in Reference [30].
Figure 5. Sensitivity analysis of the numerical simulation using MIKE 3 FM for the different mesh sizes of the experiments in Reference [30].
Figure 5. Sensitivity analysis of the numerical simulation using MIKE 3 FM for the different mesh sizes of the experiments in Reference [30].
Figure 6. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for dry-bed (α=0). The experimental data are from Reference [30].
Figure 6. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for dry-bed (α=0). The experimental data are from Reference [30].
Figure 7. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for a wet-bed (α = 0.1). The experimental data are from Reference [30].
Figure 7. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for a wet-bed (α = 0.1). The experimental data are from Reference [30].
Figure 8. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for the wet-bed (α = 0.4). The experimental data are from Reference [30].
Figure 8. Comparison between observed and simulated free surface profiles at dimensionless times T = t(g/h0)1/2 and for the wet-bed (α = 0.4). The experimental data are from Reference [30].
Figure 9. Experimental and numerical comparison of free surface profiles h/h0(x/h0) during late stages at various dimensionless times T after the failure in the dry-bed by Khankandi et al. [39].
Figure 9. Experimental and numerical comparison of free surface profiles h/h0(x/h0) during late stages at various dimensionless times T after the failure in the dry-bed by Khankandi et al. [39].

Table 2. RMSE values for the free surface profiles observed by Khankandi et al. [39].

Table 2. RMSE values for the free surface profiles observed by Khankandi et al. [39].
Table 2. RMSE values for the free surface profiles observed by Khankandi et al. [39].
Figure 10. Measured and computed water level hydrograph at various positions for dry-bed by Khankandi et al. [39]: (a) G1 (−0.5 m); (b) G2 (−0.1 m); (c) G3 (0.1 m); (d) G4 (0.8 m); (e) G6 (1.2 m); (f) G8 (5.5 m).
Figure 10. Measured and computed water level hydrograph at various positions for dry-bed by Khankandi et al. [39]: (a) G1 (−0.5 m); (b) G2 (−0.1 m); (c) G3 (0.1 m); (d) G4 (0.8 m); (e) G6 (1.2 m); (f) G8 (5.5 m).
Figure 11. Measured and computed water level hydrographs at various positions for the wet-bed by Khankandi et al. [39]: (a) G1 (−0.5 m); (b) G2 (−0.1 m); (c) G4 (0.8 m); and (d) G5 (1.0 m).
Figure 11. Measured and computed water level hydrographs at various positions for the wet-bed by Khankandi et al. [39]: (a) G1 (−0.5 m); (b) G2 (−0.1 m); (c) G4 (0.8 m); and (d) G5 (1.0 m).

Table 3. RMSE values for the water depth variations observed by Khankandi et al. [39] at the late stage.

Table 3. RMSE values for the water depth variations observed by Khankandi et al. [39] at the late stage.
Table 3. RMSE values for the water depth variations observed by Khankandi et al. [39] at the late stage.
Figure 13. Comparison of simulated velocity profiles at various locations upstream and downstream of the dam at t = 0.8 s, 2 s, and 5 s for water depth ratios α = 0.1 by Ozmen-Cagatay and Kocaman [30]: (a) P1(−1 m); (b) P3 (+0.2 m); (c) P5 (+1 m); and (d) P6 (+2 m).
Figure 13. Comparison of simulated velocity profiles at various locations upstream and downstream of the dam at t = 0.8 s, 2 s, and 5 s for water depth ratios α = 0.1 by Ozmen-Cagatay and Kocaman [30]: (a) P1(−1 m); (b) P3 (+0.2 m); (c) P5 (+1 m); and (d) P6 (+2 m).
Table 5. The required computational time for the two models to address dam break flows in all cases
Table 5. The required computational time for the two models to address dam break flows in all cases

References

  1. Gallegos, H.A.; Schubert, J.E.; Sanders, B.F. Two-dimensional high-resolution modeling of urban dam-break flooding: A case study of Baldwin Hills, California. Adv. Water Resour. 200932, 1323–1335. [Google Scholar] [CrossRef]
  2. Kim, K.S. A Mesh-Free Particle Method for Simulation of Mobile-Bed Behavior Induced by Dam Break. Appl. Sci. 20188, 1070. [Google Scholar] [CrossRef]
  3. Robb, D.M.; Vasquez, J.A. Numerical simulation of dam-break flows using depth-averaged hydrodynamic and three-dimensional CFD models. In Proceedings of the Canadian Society for Civil Engineering Hydrotechnical Conference, Québec, QC, Canada, 21–24 July 2015. [Google Scholar]
  4. LaRocque, L.A.; Imran, J.; Chaudhry, M.H. 3D numerical simulation of partial breach dam-break flow using the LES and k-ε. J. Hydraul. Res. 201351, 145–157. [Google Scholar] [CrossRef]
  5. Ritter, A. Die Fortpflanzung der Wasserwellen (The propagation of water waves). Z. Ver. Dtsch. Ing. 189236, 947–954. [Google Scholar]
  6. Dressler, R.F. Hydraulic resistance effect upon the dam-break functions. J. Res. Nat. Bur. Stand. 195249, 217–225. [Google Scholar] [CrossRef]
  7. Dressler, R.F. Comparison of theories and experiments for the hydraulic dam-break wave. Int. Assoc. Sci. Hydrol. 195438, 319–328. [Google Scholar]
  8. Whitham, G.B. The effects of hydraulic resistance in the dam-break problem. Proc. R. Soc. Lond. 1955227A, 399–407. [Google Scholar] [CrossRef]
  9. Stoker, J.J. Water Waves: The Mathematical Theory with Applications; Wiley and Sons: New York, NY, USA, 1957; ISBN 0-471-57034-6. [Google Scholar]
  10. Marshall, G.; Méndez, R. Computational Aspects of the Random Choice Method for Shallow Water Equations. J. Comput. Phys. 198139, 1–21. [Google Scholar] [CrossRef]
  11. Godunov, S.K. Finite Difference Methods for the Computation of Discontinuous Solutions of the Equations of Fluid Dynamics. Math. Sb. 195947, 271–306. [Google Scholar]
  12. Toro, E.F. Shock-Capturing Methods for Free-Surface Shallow Flows; Wiley and Sons Ltd.: New York, NY, USA, 2001. [Google Scholar]
  13. Chanson, H. Application of the method of characteristics to the dam break wave problem. J. Hydraul. Res. 200947, 41–49. [Google Scholar] [CrossRef][Green Version]
  14. Cagatay, H.; Kocaman, S. Experimental Study of Tail Water Level Effects on Dam-Break Flood Wave Propagation; 2008 Kubaba Congress Department and Travel Services: Ankara, Turkey, 2008; pp. 635–644. [Google Scholar]
  15. Stansby, P.K.; Chegini, A.; Barnes, T.C.D. The initial stages of dam-break flow. J. Fluid Mech. 1998374, 407–424. [Google Scholar] [CrossRef]
  16. Soares-Frazao, S.; Zech, Y. Dam Break in Channels with 90° Bend. J. Hydraul. Eng. 2002128, 956–968. [Google Scholar] [CrossRef]
  17. Zolghadr, M.; Hashemi, M.R.; Zomorodian, S.M.A. Assessment of MIKE21 model in dam and dike-break simulation. IJST-Trans. Mech. Eng. 201135, 247–262. [Google Scholar]
  18. Bukreev, V.I.; Gusev, A.V. Initial stage of the generation of dam-break waves. Dokl. Phys. 200550, 200–203. [Google Scholar] [CrossRef]
  19. Soares-Frazao, S.; Noel, B.; Zech, Y. Experiments of dam-break flow in the presence of obstacles. Proc. River Flow 20042, 911–918. [Google Scholar]
  20. Aureli, F.; Maranzoni, A.; Mignosa, P.; Ziveri, C. Dambreak flows: Acquisition of experimental data through an imaging technique and 2D numerical modelling. J. Hydraul. Eng. 2008134, 1089–1101. [Google Scholar] [CrossRef]
  21. Rehman, K.; Cho, Y.S. Bed Evolution under Rapidly Varying Flows by a New Method for Wave Speed Estimation. Water 20168, 212. [Google Scholar] [CrossRef]
  22. Wu, G.F.; Yang, Z.H.; Zhang, K.F.; Dong, P.; Lin, Y.T. A Non-Equilibrium Sediment Transport Model for Dam Break Flow over Moveable Bed Based on Non-Uniform Rectangular Mesh. Water 201810, 616. [Google Scholar] [CrossRef]
  23. Ferrari, A.; Fraccarollo, L.; Dumbser, M.; Toro, E.F.; Armanini, A. Three-dimensional flow evolution after a dam break. J. Fluid Mech. 2010663, 456–477. [Google Scholar] [CrossRef]
  24. Liang, D. Evaluating shallow water assumptions in dam-break flows. Proc. Inst. Civ. Eng. Water Manag. 2010163, 227–237. [Google Scholar] [CrossRef]
  25. Biscarini, C.; Francesco, S.D.; Manciola, P. CFD modelling approach for dam break flow studies. Hydrol. Earth Syst. Sci. 201014, 705–718. [Google Scholar] [CrossRef][Green Version]
  26. Oertel, M.; Bung, D.B. Initial stage of two-dimensional dam-break waves: Laboratory versus VOF. J. Hydraul. Res. 201250, 89–97. [Google Scholar] [CrossRef]
  27. Quecedo, M.; Pastor, M.; Herreros, M.I.; Merodo, J.A.F.; Zhang, Q. Comparison of two mathematical models for solving the dam break problem using the FEM method. Comput. Method Appl. Mech. Eng. 2005194, 3984–4005. [Google Scholar] [CrossRef]
  28. Shigematsu, T.; Liu, P.L.F.; Oda, K. Numerical modeling of the initial stages of dam-break waves. J. Hydraul. Res. 200442, 183–195. [Google Scholar] [CrossRef]
  29. Soares-Frazao, S. Experiments of dam-break wave over a triangular bottom sill. J. Hydraul. Res. 200745, 19–26. [Google Scholar] [CrossRef]
  30. Ozmen-Cagatay, H.; Kocaman, S. Dam-break flows during initial stage using SWE and RANS approaches. J. Hydraul. Res. 201048, 603–611. [Google Scholar] [CrossRef]
  31. Vasquez, J.; Roncal, J. Testing River2D and FLOW-3D for Sudden Dam-Break Flow Simulations. In Proceedings of the Canadian Dam Association’s 2009 Annual Conference: Protecting People, Property and the Environment, Whistler, BC, Canada, 3–8 October 2009. [Google Scholar]
  32. Ozmen-Cagatay, H.; Kocaman, S. Dam-break flow in the presence of obstacle: Experiment and CFD simulation. Eng. Appl. Comput. Fluid 20115, 541–552. [Google Scholar] [CrossRef]
  33. Ozmen-Cagatay, H.; Kocaman, S.; Guzel, H. Investigation of dam-break flood waves in a dry channel with a hump. J. Hydro-Environ. Res. 20148, 304–315. [Google Scholar] [CrossRef]
  34. Gu, S.L.; Zheng, S.P.; Ren, L.Q.; Xie, H.W.; Huang, Y.F.; Wei, J.H.; Shao, S.D. SWE-SPHysics Simulation of Dam Break Flows at South-Gate Gorges Reservoir. Water 20179, 387. [Google Scholar] [CrossRef]
  35. Evangelista, S. Experiments and Numerical Simulations of Dike Erosion due to a Wave Impact. Water 20157, 5831–5848. [Google Scholar] [CrossRef][Green Version]
  36. Bocci, M.; Chiarlo, R.; De Nat, L.; Fanelli, A.; Petersen, O.; Sorensen, J.T.; Friss-Christensen, A. Modelling of impacts from a long sea outfall outside of the Venice Lagoon (Italy). In Proceedings of the MWWD—IEMES 2006 Conference, Antalya, Turkey, 6–10 November 2006; MWWD Organization: Antalya, Turkey, 2006. [Google Scholar]
  37. Nikolaos, T.F.; Georgios, M.H. Three-dimensional numerical simulation of wind-induced barotropic circulation in the Gulf of Patras. Ocean Eng. 201037, 355–364. [Google Scholar]
  38. Goyal, R.; Rathod, P. Hydrodynamic Modelling for Salinity of Singapore Strait and Johor Strait using MIKE 3FM. In Proceedings of the 2011 2nd International Conference on Environmental Science and Development, Singapore, 26–28 February 2011. [Google Scholar]
  39. Khankandi, A.F.; Tahershamsi, A.; Soares-Frazão, S. Experimental investigation of reservoir geometry effect on dam-break flow. J. Hydraul. Res. 201250, 376–387. [Google Scholar] [CrossRef]
  40. Flow Science Inc. FLOW-3D User’s Manuals; Flow Science Inc.: Santa Fe, NM, USA, 2007. [Google Scholar]
  41. Danish Hydraulic Institute (DHI). MIKE 3 Flow Model FM. Hydrodynamic Module-User Guide; DHI: Horsholm, Denmark, 2014. [Google Scholar]
  42. Pilotti, M.; Tomirotti, M.; Valerio, G. Simplified Method for the Characterization of the Hydrograph following a Sudden Partial Dam Break. J. Hydraul. Eng. 2010136, 693–704. [Google Scholar] [CrossRef]
  43. Hooshyaripor, F.; Tahershamsi, A.; Razi, S. Dam break flood wave under different reservoir’s capacities and lengths. Sādhanā 201742, 1557–1569. [Google Scholar] [CrossRef]
  44. Kocaman, S.; Ozmen-Cagatay, H. Investigation of dam-break induced shock waves impact on a vertical Wall. J. Hydrol. 2015525, 1–12. [Google Scholar] [CrossRef]
  45. Liu, H.; Liu, H.J.; Guo, L.H.; Lu, S.X. Experimental Study on the Dam-Break Hydrographs at the Gate Location. J. Ocean Univ. China 201716, 697–702. [Google Scholar] [CrossRef]
  46. Marra, D.; Earl, T.; Ancey, C. Experimental Investigations of Dam Break Flows down an Inclined Channel. In Proceedings of the 34th World Congress of the International Association for Hydro- Environment Research and Engineering: 33rd Hydrology and Water Resources Symposium and 10th Conference on Hydraulics in Water Engineering, Brisbane, Australia, 26 June–1 July 2011. [Google Scholar]
  47. Wang, J.; Liang, D.F.; Zhang, J.X.; Xiao, Y. Comparison between shallow water and Boussinesq models for predicting cascading dam-break flows. Nat. Hazards 201683, 327–343. [Google Scholar] [CrossRef]
  48. Yang, C.; Lin, B.L.; Jiang, C.B.; Liu, Y. Predicting near-field dam-break flow and impact force using a 3D model. J. Hydraul. Res. 201048, 784–792. [Google Scholar] [CrossRef]
Figure 5 - Modeling a simple lotus overflow symmetrically in FLOW-3D software

Flow-3D를 이용한 나팔형 여수로의 방류계수에 대한 와류방지 블레이드 45 도의 효과

Effect of Vortex Breaker Blades 45 Degree on Discharge Coefficient of Morning Glory Spillway Using Flow-3D

Authors

S. Noruzi1
and J. Ahadiyan2*
1– M.Sc. Student, Faculty of Water Sciences Engineering, Shahid Chamran University of Ahvaz, Iran.
2*-Corresponding Author, Associate Professor, Faculty of Water Sciences Engineering, Shahid Chamran
University of Ahvaz, Iran.

Abstract

The discharge coefficient of morning glory spillway is decreased with eddies created by vortex at the inlet part of weir. However, a series of specific blades can reduce vortices which result in the spillway efficiency is increased. Hence, in this research numerical modeling of installed breaker blade on morning glory spillway was evaluated using Flow-3D model. To achieve these purposes, morning glory spillway was modeled without and with blades 3, 4 and 6 blades at 45 degree angle. To simulate the turbulence fluctuations, the modified k-e model (RNG k-e) was used and its results were compared to the experimental data. Results showed that by installing blades, the discharge coefficient increases up to 42 percent with 25 percent decreasing in the upstream water level. Moreover, among the three different arrangements of blades, the six-blade model was found to have more satisfactory results than other models. In comparison to control model, for H/D between 0 to 0.1 and 0.1 to 0.2 the discharge coefficient has been increased 40 and 57 percent for six-blade arrangement, respectively. 

모닝 글로리의 방류계수는 위어 입구 부분의 와류에 의해 생성된 소용돌이로 감소합니다. 그러나 일련의 특정 블레이드는 와류를 줄여 여수로 효율성을 높일 수 있습니다. 따라서 본 연구에서는 모닝 글로리 여수로에 설치된 브레이커 블레이드의 수치 모델링을 FLOW-3D 모델을 사용하여 평가했습니다. 이러한 목적을 달성하기 위해 모닝 글로리 여수로는 45도 각도로 블레이드 3, 4 및 6 블레이드 있는 모델과 없는 모델로 모델링되었습니다. 난류 변동을 시뮬레이션하기 위해 수정된 k-e 모델 (RNG k-e)을 사용하고 그 결과를 실험 데이터와 비교했습니다. 결과에 따르면 블레이드를 설치하면 상류 수위가 25 % 감소하면서 배출 계수가 42 %까지 증가합니다. 또한 3 개의 서로다른 블레이드 배열 중 6 개 블레이드 모델이 다른 모델보다 더 만족스러운 결과를 나타냈다. 기본 모델과 비교하여 H / D가 0 ~ 0.1 및 0.1 ~ 0.2 인 경우, 6개 블레이드 배열에서 방류계수가 각각 40 % 및 57 % 증가했습니다.

Keywords

Figure 1 - Dimensions of the vortex blade
Figure 1 – Dimensions of the vortex blade
Figure 3 - A (Physical model of lotus overflow without blade, b) Physical model of lotus overflow with eddy blades.
Figure 3 – A (Physical model of lotus overflow without blade, b) Physical model of lotus overflow with eddy blades.
Figure 5 - Modeling a simple lotus overflow symmetrically in FLOW-3D software
Figure 5 – Modeling a simple lotus overflow symmetrically in FLOW-3D software
Figure 7 - Comparison of Ashley flow chart with numerical model and laboratory
Figure 7 – Comparison of Ashley flow chart with numerical model and laboratory
Figure 8 - Comparison of flow coefficient diagram - immersion ratio of numerical model with laboratory: a (overflow without blade, b) overflow with three blades, c (overflow with four blades, d) overflow with six blades
Figure 8 – Comparison of flow coefficient diagram – immersion ratio of numerical model with laboratory: a (overflow without blade, b) overflow with three blades, c (overflow with four blades, d) overflow with six blades

Reference

1 -حیدری ارجلو، س.، موسوی جهرمی، س. ح. و ادیب، ا. 1386 .بررسی تاثیر شیب بر تعداد بهینه پلکانها در سرریزهای پلکانی، مجله علوم و مهندسی
.)123-136 :)2(33 ،كشاورزی علمی )آبیاری
2 -حاجیپور، گ. 1363 .بررسی آزمایشگاهی تأثیر تیغههای گردابشکن بر هیدرولیک جریان سرریز نیلوفری. پایاننامه كارشناسی ارشد رشته سازههای آبی،
دانشکده مهندسی علوم آب، دانشگاه شهید چمران اهواز.
3 -رنجبر ملکشاه، م.، 1363 .بررسی رفتار سرریز نیلوفری با پایین دست تاج پلکانی بوسیله مدلسازی رایانهای، پایاننامه كارشناسی ارشد مهندسی عمران،
دانشکده مهندسی عمران، دانشگاه خواجه نصیر طوسی.
4 -رمضانی، س. كاویانپور، م ر. و ع. حسنی نژاد. 1362 .بررسی پارامترهای مؤثر بر آبگذری سرریزهای نیلوفری. هفتمین كنگره ملی مهندسی عمران،
دانشکده مهندسی شهید نیکبخت، زاهدان.
1 -سامانی، م. 1331 .طراحی سازههای هیدرولیکی. انتشارات شركت مهندسی مشاور دز آب اهواز
1 -قاسمزاده، ف. 1362 .شبیه سازی مسائل هیدرولیکی در 3D-FLOW .تهران، نوآور.
3 -كمانبدست، 1 ،.موسوی، س.ر. 1361 .مطالعه آزمایشگاهی تأثیر تعداد و زاویه گرداب شکن بر مشخصات جریان در سرریز نیلوفری مربعی، نشریه علوم
آب و خاک )غعلوم و فنون كشاورزی و منابع طبیعی(، سال بیستم، شماره 38 ،صفحه 182-131 .
8 -نظری پوركیانی، ع ا. 1363 .بررسی فشار و سرعت جریان در سرریز نیلوفری سد البرز با استفاده از نرمافزار 3D-FLOW .اولین كنفرانس سراسری
توسعه محوری مهندسی عمران، معماری، برق و مکانیک ایران.
6 -نوحانی، ا.، جمالی امام قیس، ر. 1364 .بررسی آزمایشگاهی تأثیرشکل تیغه های ضد گرداب برراندمان تخلیهی سرریزهای نیلوفری، نشریه آبیاری و
زهکشی ایران، جلد 6 ،شماره 1 ،صفحه 346-341 .
10-Akbari, A A., Nohani, E and A. Afrous. 2015. Numerical study of the effect of anti-vortex plates on the
inflow pattern in shaft spillways. Indian Journal of Fundamental and Applied Life Sciences, 5(S1):
3819-3826.
11-Anonymous, 1965. Design of Small Dams. Water Resources Technical publication, U.S Department of
the interior Bureau of Reclamation.
12-Bagheri, A., Shafai Bajestan, M., Mousavi Jahromi, H., Kashkuli, H. and H. Sedghi. 2010. Hydraulic
evaluation of the flow over polyhedral morning glory spillways. Word Applied Sciences Journal, 9(7):
712-717.
13- Fattor, C. A. and J. D. Bacchiega. 2003. Analysis of instabilities in the charge of regime in morning
glory spillways. Journal of Hydraulic Research, 40(4): 114-123.
14- Khatsuria, R. M. 2005. Hydraulics of spillways and energy dissipaters. Marcel Dekker. Department of
Civil and Environmental Engineering Georgia, Institute of Technology Atlanta, Newyork, USA.
15-Mousavi. S. R., Kamanbedast, A.A., and H. Fathian. 2013. Experimental investigation of the effect of
number of anti-vortex piers on submergence threshold in morning glory spillway with square inlet.
Technical Journal of Engineering and Applied Sciences, 3(24): 3534-3540.
16- Novak, P. 2007. Hydraulic Structures, Fourth edition published by Taylor and Francis. University of
New Castle upon, Tyne, UK, Landon and Network.
17-Tavana, M H., Mousavi Jahromi, H., Shafai Bajestan, M., Masjedi, A. R. and H. Sedghi. 2011.
Optimization of number and direction of vortex breakers in the morning glory spillway using physical
model. Economy, Environmental and Conservation Journal, 17(2): 435-440.
18-Vresteeg. H. K and W. Malalasekera. 1995. An introduction to computational fluid dynamics. Longman
Scientific and Technical. New York.
19-Yakhot. V and L. M. Smith. 1992. The renormalization group. The e-expansion and of turbulence
models. Journal of Computing, 7(1): 35-61.

The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

Numerical investigation of flow characteristics over stepped spillways

Güven, Aytaç
Mahmood, Ahmed Hussein
Water Supply (2021) 21 (3): 1344–1355.
https://doi.org/10.2166/ws.2020.283Article history

Abstract

Spillways are constructed to evacuate flood discharge safely so that a flood wave does not overtop the dam body. There are different types of spillways, with the ogee type being the conventional one. A stepped spillway is an example of a nonconventional spillway. The turbulent flow over a stepped spillway was studied numerically by using the Flow-3D package. Different fluid flow characteristics such as longitudinal flow velocity, temperature distribution, density and chemical concentration can be well simulated by Flow-3D. In this study, the influence of slope changes on flow characteristics such as air entrainment, velocity distribution and dynamic pressures distribution over a stepped spillway was modelled by Flow-3D. The results from the numerical model were compared with an experimental study done by others in the literature. Two models of a stepped spillway with different discharge for each model were simulated. The turbulent flow in the experimental model was simulated by the Renormalized Group (RNG) turbulence scheme in the numerical model. A good agreement was achieved between the numerical results and the observed ones, which are exhibited in terms of graphics and statistical tables.

배수로는 홍수가 댐 몸체 위로 넘치지 않도록 안전하게 홍수를 피할 수 있도록 건설되었습니다. 다른 유형의 배수로가 있으며, ogee 유형이 기존 유형입니다. 계단식 배수로는 비 전통적인 배수로의 예입니다. 계단식 배수로 위의 난류는 Flow-3D 패키지를 사용하여 수치적으로 연구되었습니다.

세로 유속, 온도 분포, 밀도 및 화학 농도와 같은 다양한 유체 흐름 특성은 Flow-3D로 잘 시뮬레이션 할 수 있습니다. 이 연구에서는 계단식 배수로에 대한 공기 혼입, 속도 분포 및 동적 압력 분포와 같은 유동 특성에 대한 경사 변화의 영향을 Flow-3D로 모델링 했습니다.

수치 모델의 결과는 문헌에서 다른 사람들이 수행한 실험 연구와 비교되었습니다. 각 모델에 대해 서로 다른 배출이 있는 계단식 배수로의 두 모델이 시뮬레이션되었습니다. 실험 모델의 난류 흐름은 수치 모델의 Renormalized Group (RNG) 난류 계획에 의해 시뮬레이션되었습니다. 수치 결과와 관찰 된 결과 사이에 좋은 일치가 이루어졌으며, 이는 그래픽 및 통계 테이블로 표시됩니다.

HIGHLIGHTS

ListenReadSpeaker webReader: Listen

  • A numerical model was developed for stepped spillways.
  • The turbulent flow was simulated by the Renormalized Group (RNG) model.
  • Both numerical and experimental results showed that flow characteristics are greatly affected by abrupt slope change on the steps.

Keyword

CFDnumerical modellingslope changestepped spillwayturbulent flow

INTRODUCTION

댐 구조는 물 보호가 생활의 핵심이기 때문에 물을 저장하거나 물을 운반하는 전 세계에서 가장 중요한 프로젝트입니다. 그리고 여수로는 댐의 가장 중요한 부분 중 하나로 분류됩니다. 홍수로 인한 파괴 나 피해로부터 댐을 보호하기 위해 여수로가 건설됩니다.

수력 발전, 항해, 레크리에이션 및 어업의 중요성을 감안할 때 댐 건설 및 홍수 통제는 전 세계적으로 매우 중요한 문제로 간주 될 수 있습니다. 많은 유형의 배수로가 있지만 가장 일반적인 유형은 다음과 같습니다 : ogee 배수로, 자유 낙하 배수로, 사이펀 배수로, 슈트 배수로, 측면 채널 배수로, 터널 배수로, 샤프트 배수로 및 계단식 배수로.

그리고 모든 여수로는 입구 채널, 제어 구조, 배출 캐리어 및 출구 채널의 네 가지 필수 구성 요소로 구성됩니다. 특히 롤러 압축 콘크리트 (RCC) 댐 건설 기술과 더 쉽고 빠르며 저렴한 건설 기술로 분류 된 계단식 배수로 건설과 관련하여 최근 수십 년 동안 많은 계단식 배수로가 건설되었습니다 (Chanson 2002; Felder & Chanson 2011).

계단식 배수로 구조는 캐비테이션 위험을 감소시키는 에너지 소산 속도를 증가시킵니다 (Boes & Hager 2003b). 계단식 배수로는 다양한 조건에서 더 매력적으로 만드는 장점이 있습니다.

계단식 배수로의 흐름 거동은 일반적으로 낮잠, 천이 및 스키밍 흐름 체제의 세 가지 다른 영역으로 분류됩니다 (Chanson 2002). 유속이 낮을 때 nappe 흐름 체제가 발생하고 자유 낙하하는 낮잠의 시퀀스로 특징 지워지는 반면, 스키밍 흐름 체제에서는 물이 외부 계단 가장자리 위의 유사 바닥에서 일관된 흐름으로 계단 위로 흐릅니다.

또한 주요 흐름에서 3 차원 재순환 소용돌이가 발생한다는 것도 분명합니다 (예 : Chanson 2002; Gonzalez & Chanson 2008). 계단 가장자리 근처의 의사 바닥에서 흐름의 방향은 가상 바닥과 가상으로 정렬됩니다. Takahashi & Ohtsu (2012)에 따르면, 스키밍 흐름 체제에서 주어진 유속에 대해 흐름은 계단 가장자리 근처의 수평 계단면에 영향을 미치고 슈트 경사가 감소하면 충돌 영역의 면적이 증가합니다. 전이 흐름 체제는 나페 흐름과 스키밍 흐름 체제 사이에서 발생합니다. 계단식 배수로를 설계 할 때 스키밍 흐름 체계를 고려해야합니다 (예 : Chanson 1994, Matos 2000, Chanson 2002, Boes & Hager 2003a).

CFD (Computational Fluid Dynamics), 즉 수력 공학의 수치 모델은 일반적으로 물리적 모델에 소요되는 총 비용과 시간을 줄여줍니다. 따라서 수치 모델은 실험 모델보다 빠르고 저렴한 것으로 분류되며 동시에 하나 이상의 목적으로 사용될 수도 있습니다. 사용 가능한 많은 CFD 소프트웨어 패키지가 있지만 가장 널리 사용되는 것은 FLOW-3D입니다. 이 연구에서는 Flow 3D 소프트웨어를 사용하여 유량이 서로 다른 두 모델에 대해 계단식 배수로에서 공기 농도, 속도 분포 및 동적 압력 분포를 시뮬레이션합니다.

Roshan et al. (2010)은 서로 다른 수의 계단 및 배출을 가진 계단식 배수로의 두 가지 물리적 모델에 대한 흐름 체제 및 에너지 소산 조사를 연구했습니다. 실험 모델의 기울기는 각각 19.2 %, 12 단계와 23 단계의 수입니다. 결과는 23 단계 물리적 모델에서 관찰 된 흐름 영역이 12 단계 모델보다 더 수용 가능한 것으로 간주되었음을 보여줍니다. 그러나 12 단계 모델의 에너지 손실은 23 단계 모델보다 더 많았습니다. 그리고 실험은 스키밍 흐름 체제에서 23 단계 모델의 에너지 소산이 12 단계 모델보다 약 12 ​​% 더 적다는 것을 관찰했습니다.

Ghaderi et al. (2020a)는 계단 크기와 유속이 다른 정련 매개 변수의 영향을 조사하기 위해 계단식 배수로에 대한 실험 연구를 수행했습니다. 그 결과, 흐름 체계가 냅페 흐름 체계에서 발생하는 최소 scouring 깊이와 같은 scouring 구멍 치수에 영향을 미친다는 것을 보여주었습니다. 또한 테일 워터 깊이와 계단 크기는 최대 scouring깊이에 대한 실제 매개 변수입니다. 테일 워터의 깊이를 6.31cm에서 8.54 및 11.82cm로 늘림으로써 수세 깊이가 각각 18.56 % 및 11.42 % 증가했습니다. 또한 이 증가하는 테일 워터 깊이는 scouring 길이를 각각 31.43 % 및 16.55 % 감소 시킵니다. 또한 유속을 높이면 Froude 수가 증가하고 흐름의 운동량이 증가하면 scouring이 촉진됩니다. 또한 결과는 중간의 scouring이 횡단면의 측벽보다 적다는 것을 나타냅니다. 계단식 배수로 하류의 최대 scouring 깊이를 예측 한 후 실험 결과와 비교하기 위한 실험식이 제안 되었습니다. 그리고 비교 결과 제안 된 공식은 각각 3.86 %와 9.31 %의 상대 오차와 최대 오차 내에서 scouring 깊이를 예측할 수 있음을 보여주었습니다.

Ghaderi et al. (2020b)는 사다리꼴 미로 모양 (TLS) 단계의 수치 조사를 했습니다. 결과는 이러한 유형의 배수로가 확대 비율 LT / Wt (LT는 총 가장자리 길이, Wt는 배수로의 폭)를 증가시키기 때문에 더 나은 성능을 갖는 것으로 관찰되었습니다. 또한 사다리꼴 미로 모양의 계단식 배수로는 더 큰 마찰 계수와 더 낮은 잔류 수두를 가지고 있습니다. 마찰 계수는 다양한 배율에 대해 0.79에서 1.33까지 다르며 평평한 계단식 배수로의 경우 대략 0.66과 같습니다. 또한 TLS 계단식 배수로에서 잔류 수두의 비율 (Hres / dc)은 약 2.89이고 평평한 계단식 배수로의 경우 약 4.32와 같습니다.

Shahheydari et al. (2015)는 Flow-3D 소프트웨어, RNG k-ε 모델 및 VOF (Volume of Fluid) 방법을 사용하여 배출 계수 및 에너지 소산과 같은 자유 표면 흐름의 프로파일을 연구하여 스키밍 흐름 체제에서 계단식 배수로에 대한 흐름을 조사했습니다. 실험 결과와 비교했습니다. 결과는 에너지 소산 율과 방전 계수율의 관계가 역으로 실험 모델의 결과와 잘 일치 함을 보여 주었다.

Mohammad Rezapour Tabari & Tavakoli (2016)는 계단 높이 (h), 계단 길이 (L), 계단 수 (Ns) 및 단위 폭의 방전 (q)과 같은 다양한 매개 변수가 계단식 에너지 ​​소산에 미치는 영향을 조사했습니다. 방수로. 그들은 해석에 FLOW-3D 소프트웨어를 사용하여 계단식 배수로에서 에너지 손실과 임계 흐름 깊이 사이의 관계를 평가했습니다. 또한 유동 난류에 사용되는 방정식과 표준 k-ɛ 모델을 풀기 위해 유한 체적 방법을 적용했습니다. 결과에 따르면 스텝 수가 증가하고 유량 배출량이 증가하면 에너지 손실이 감소합니다. 얻은 결과를 다른 연구와 비교하고 경험적, 수학적 조사를 수행하여 결국 합격 가능한 결과를 얻었습니다.

METHODOLOGY

ListenReadSpeaker webReader: ListenFor all numerical models the basic principle is very similar: a set of partial differential equations (PDE) present the physical problems. The flow of fluids (gas and liquid) are governed by the conservation laws of mass, momentum and energy. For Computational Fluid Dynamics (CFD), the PDE system is substituted by a set of algebraic equations which can be worked out by using numerical methods (Versteeg & Malalasekera 2007). Flow-3D uses the finite volume approach to solve the Reynolds Averaged Navier-Stokes (RANS) equation, by applying the technique of Fractional Area/Volume Obstacle Representation (FAVOR) to define an obstacle (Flow Science Inc. 2012). Equations (1) and (2) are RANS and continuity equations with FAVOR variables that are applied for incompressible flows.

formula

(1)

formula

(2)where  is the velocity in xi direction, t is the time,  is the fractional area open to flow in the subscript directions,  is the volume fraction of fluid in each cell, p is the hydrostatic pressure,  is the density, is the gravitational force in subscript directions and  is the Reynolds stresses.

Turbulence modelling is one of three key elements in CFD (Gunal 1996). There are many types of turbulence models, but the most common are Zero-equation models, One-equation models, Two-equation models, Reynolds Stress/Flux models and Algebraic Stress/Flux models. In FLOW-3D software, five turbulence models are available. The formulation used in the FLOW-3D software differs slightly from other formulations that includes the influence of the fractional areas/volumes of the FAVORTM method and generalizes the turbulence production (or decay) associated with buoyancy forces. The latter generalization, for example, includes buoyancy effects associated with non-inertial accelerations.

The available turbulence models in Flow-3D software are the Prandtl Mixing Length Model, the One-Equation Turbulent Energy Model, the Two-Equation Standard  Model, the Two-Equation Renormalization-Group (RNG) Model and large Eddy Simulation Model (Flow Science Inc. 2012).In this research the RNG model was selected because this model is more commonly used than other models in dealing with particles; moreover, it is more accurate to work with air entrainment and other particles. In general, the RNG model is classified as a more widely-used application than the standard k-ɛ model. And in particular, the RNG model is more accurate in flows that have strong shear regions than the standard k-ɛ model and it is defined to describe low intensity turbulent flows. For the turbulent dissipation  it solves an additional transport equation:

formula

(3)where CDIS1, CDIS2, and CDIS3 are dimensionless parameters and the user can modify them. The diffusion of dissipation, Diff ɛ, is

formula

(4)where uv and w are the x, y and z coordinates of the fluid velocity; ⁠, ⁠,  and ⁠, are FLOW-3D’s FAVORTM defined terms;  and  are turbulence due to shearing and buoyancy effects, respectively. R and  are related to the cylindrical coordinate system. The default values of RMTKE, CDIS1 and CNU differ, being 1.39, 1.42 and 0.085 respectively. And CDIS2 is calculated from turbulent production (⁠⁠) and turbulent kinetic energy (⁠⁠).The kinematic turbulent viscosity is the same in all turbulence transport models and is calculated from

formula

(5)where ⁠: is the turbulent kinematic viscosity.  is defined as the numerical challenge between the RNG and the two-equation k-ɛ models, found in the equation below. To avoid an unphysically large result for  in Equation (3), since this equation could produce a value for  very close to zero and also because the physical value of  may approach to zero in such cases, the value of  is calculated from the following equation:

formula

(6)where ⁠: the turbulent length scale.

VOF and FAVOR are classifications of volume-fraction methods. In these two methods, firstly the area should be subdivided into a control volume grid or a small element. Each flow parameter like velocity, temperature and pressure values within the element are computed for each element containing liquids. Generally, these values represent the volumetric average of values in the elements.Numerous methods have been used recently to solve free infinite boundaries in the various numerical simulations. VOF is an easy and powerful method created based on the concept of a fractional intensity of fluid. A significant number of studies have confirmed that this method is more flexible and efficient than others dealing with the configurations of a complex free boundary. By using VOF technology the Flow-3D free surface was modelled and first declared in Hirt & Nichols (1981). In the VOF method there are three ingredients: a planner to define the surface, an algorithm for tracking the surface as a net mediator moving over a computational grid, and application of the boundary conditions to the surface. Configurations of the fluids are defined in terms of VOF function, F (x, y, z, t) (Hirt & Nichols 1981). And this VOF function shows the volume of flow per unit volume

formula

(7)

formula

(8)

formula

(9)where  is the density of the fluid, is a turbulent diffusion term,  is a mass source,  is the fractional volume open to flow. The components of velocity (u, v, w) are in the direction of coordinates (x, y, z) or (r, ⁠).  in the x-direction is the fractional area open to flow,  and  are identical area fractions for flow in the y and z directions. The R coefficient is based on the selection of the coordinate system.

The FAVOR method is a different method and uses another volume fraction technique, which is only used to define the geometry, such as the volume of liquid in each cell used to determine the position of fluid surfaces. Another fractional volume can be used to define the solid surface. Then, this information is used to determine the boundary conditions of the wall that the flow should be adapted for.

Case study

ListenReadSpeaker webReader: Listen

In this study, the experimental results of Ostad Mirza (2016) was simulated. In a channel composed of two 4 m long modules, with a transparent sidewall of height 0.6 m and 0.5 m width. The upstream chute slope (i.e. pseudo-bottom angle) Ɵ1 = 50°, the downstream chute slope Ɵ2 = 30° or 18.6°, the step heights h = 0.06 m, the total number of steps along the 50° chute 41 steps, the total number of steps along the 30° chute 34 steps and the total number of steps along the 18.6° chute 20 steps.

The flume inflow tool contained a jetbox with a maximum opening set to 0.12 meters, designed for passing the maximum unit discharge of 0.48 m2/s. The measurements of the flow properties (i.e. air concentration and velocity) were computed perpendicular to the pseudo-bottom as shown in Figure 1 at the centre of twenty stream-wise cross-sections, along the stepped chute, (i.e. in five steps up on the slope change and fifteen steps down on the slope change, namely from step number −09 to +23 on 50°–30° slope change, or from −09 to +15 on 50°–18.6° slope change, respectively).

Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).
Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).

Sketch of the air concentration C and velocity V measured perpendicular to the pseudo-bottom used by Mirza (Ostad Mirza 2016).

Pressure sensors were arranged with the x/l values for different slope change as shown in Table 1, where x is the distance from the step edge, along the horizontal step face, and l is the length of the horizontal step face. The location of pressure sensors is shown in Table 1.Table 1

Location of pressure sensors on horizontal step faces

Θ(°)L(m)x/l (–)
50.0 0.050 0.35 0.64 – – – 
30.0 0.104 0.17 0.50 0.84 – – 
18.6 0.178 0.10 0.30 0.50 0.7 0.88 
Location of pressure sensors on horizontal step faces
Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.
Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.

Inlet boundary condition for Q = 0.235 m3/s and fluid elevation 4.21834 m.

Numerical model set-up

ListenReadSpeaker webReader: Listen

A 3D numerical model of hydraulic phenomena was simulated based on an experimental study by Ostad Mirza (2016). The water surcharge and flow pressure over the stepped spillway was computed for two models of a stepped spillway with different discharge for each model. In this study, the package was used to simulate the flow parameters such as air entrainment, velocity distribution and dynamic pressures. The solver uses the finite volume technique to discretize the computational domain. In every test run, one incompressible fluid flow with a free surface flow selected at 20̊ was used for this simulation model. Table 2 shows the variables used in test runs.Table 2

Variables used in test runs

Test no.Θ1 (°)Θ2 (°)h(m)d0q (m3s1)dc/h (–)
50 18.6 0.06 0.045 0.1 2.6 
50 18.6 0.06 0.082 0.235 4.6 
50 30.0 0.06 0.045 0.1 2.6 
50 30.0 0.06 0.082 0.235 4.6 
Table 2 Variables used in test runs

For stepped spillway simulation, several parameters should be specified to get accurate simulations, which is the scope of this research. Viscosity and turbulent, gravity and non-inertial reference frame, air entrainment, density evaluation and drift-flux should be activated for these simulations. There are five different choices in the ‘viscosity and turbulent’ option, in the viscosity flow and Renormalized Group (RNG) model. Then a dynamical model is selected as the second option, the ‘gravity and non-inertial reference frame’. Only the z-component was inputted as a negative 9.81 m/s2 and this value represents gravitational acceleration but in the same option the x and y components will be zero. Air entrainment is selected. Finally, in the drift-flux model, the density of phase one is input as (water) 1,000 kg/m3 and the density of phase two (air) as 1.225 kg/m3. Minimum volume fraction of phase one is input equal to 0.1 and maximum volume fraction of phase two to 1 to allow air concentration to reach 90%, then the option allowing gas to escape at free surface is selected, to obtain closer simulation.

The flow domain is divided into small regions relatively by the mesh in Flow-3D numerical model. Cells are the smallest part of the mesh, in which flow characteristics such as air concentration, velocity and dynamic pressure are calculated. The accuracy of the results and simulation time depends directly on the mesh block size so the cell size is very important. Orthogonal mesh was used in cartesian coordinate systems. A smaller cell size provides more accuracy for results, so we reduced the number of cells whilst including enough accuracy. In this study, the size of cells in x, y and z directions was selected as 0.015 m after several trials.

Figure 3 shows the 3D computational domain model 50–18.6 slope change, that is 6.0 m length, 0.50 m width and 4.23 m height. The 3D model of the computational domain model 50–30 slope changes this to 6.0 m length, 0.50 m width and 5.068 m height and the size of meshes in x, y, and z directions are 0.015 m. For the 50–18.6 slope change model: both total number of active and passive cells = 4,009,952, total number of active cells = 3,352,307, include real cells (used for solving the flow equations) = 3,316,269, open real cells = 3,316,269, fully blocked real cells equal to zero, external boundary cells were 36,038, inter-block boundary cells = 0 (Flow-3D report). For 50–30 slope change model: both total number of active and passive cells = 4,760,002, total number of active cells equal to 4,272,109, including real cells (used for solving the flow equations) were 3,990,878, open real cells = 3,990,878 fully blocked real cells = zero, external boundary cells were 281,231, inter-block boundary cells = 0 (Flow-3D report).

The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.
Figure3 The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

Figure 3VIEW LARGEDOWNLOAD SLIDE

The 3D computational domain model (50–18.6) slope change, and boundary condition for (50–30 slope change) model.

When solving the Navier-Stokes equation and continuous equations, boundary conditions should be applied. The most important work of boundary conditions is to create flow conditions similar to physical status. The Flow-3D software has many types of boundary condition; each type can be used for the specific condition of the models. The boundary conditions in Flow-3D are symmetry, continuative, specific pressure, grid overlay, wave, wall, periodic, specific velocity, outflow, and volume flow rate.

There are two options to input finite flow rate in the Flow-3D software either for inlet discharge of the system or for the outlet discharge of the domain: specified velocity and volume flow rate. In this research, the X-minimum boundary condition, volume flow rate, has been chosen. For X-maximum boundary condition, outflow was selected because there is nothing to be calculated at the end of the flume. The volume flow rate and the elevation of surface water was set for Q = 0.1 and 0.235 m3/s respectively (Figure 2).

The bottom (Z-min) is prepared as a wall boundary condition and the top (Z-max) is computed as a pressure boundary condition, and for both (Y-min) and (Y-max) as symmetry.

RESULTS AND DISCUSSION

ListenReadSpeaker webReader: Listen

The air concentration distribution profiles in two models of stepped spillway were obtained at an acquisition time equal to 25 seconds in skimming flow for both upstream and downstream of a slope change 50°–18.6° and 50°–30° for different discharge as in Table 2, and as shown in Figure 4 for 50°–18.6° slope change and Figure 5 for 50°–30° slope change configuration for dc/h = 4.6. The simulation results of the air concentration are very close to the experimental results in all curves and fairly close to that predicted by the advection-diffusion model for the air bubbles suggested by Chanson (1997) on a constant sloping chute.

Figure 4 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6. VIEW LARGEDOWNLOAD SLIDE Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.
Figure 4 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6. VIEW LARGEDOWNLOAD SLIDE Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.

Figure 4VIEW LARGEDOWNLOAD SLIDE

Experimental and simulated air concentration distribution for steps number −5, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 4.6.

Figure5 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.
Figure5 Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.

Figure 5VIEW LARGEDOWNLOAD SLIDE

Experimental and simulated air concentration distribution for steps number −5, +1, +5, +11, +19 and +22 along the 50°–30° slope change, for dc/h = 4.6.

Figure 6VIEW LARGEDOWNLOAD SLIDE

Figure 6 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.
Figure 6 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.

Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5, +8, +11 and +15 along the 50°–18.6° slope change for dc/h = 2.6.

Figure 7 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.
Figure 7 Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.

Figure 7VIEW LARGEDOWNLOAD SLIDE

Experimental and simulated dimensionless velocity distribution for steps number −5, −1, +1, +5. +11, +15 and +22 along the 50°–30° slope change for dc/h = 2.6.

But as is shown in all above mentioned figures it is clear that at the pseudo-bottom the CFD results of air concentration are less than experimental ones until the depth of water reaches a quarter of the total depth of water. Also the direction of the curves are parallel to each other when going up towards the surface water and are incorporated approximately near the surface water. For all curves, the cross-section is separate between upstream and downstream steps. Therefore the (-) sign for steps represents a step upstream of the slope change cross-section and the (+) sign represents a step downstream of the slope change cross-section.

The dimensionless velocity distribution (V/V90) profile was acquired at an acquisition time equal to 25 seconds in skimming flow of the upstream and downstream slope change for both 50°–18.6° and 50°–30° slope change. The simulation results are compared with the experimental ones showing that for all curves there is close similarity for each point between the observed and experimental results. The curves increase parallel to each other and they merge near at the surface water as shown in Figure 6 for slope change 50°–18.6° configuration and Figure 7 for slope change 50°–30° configuration. However, at step numbers +1 and +5 in Figure 7 there are few differences between the simulated and observed results, namely the simulation curves ascend regularly meaning the velocity increases regularly from the pseudo-bottom up to the surface water.

Figure 8 (50°–18.6° slope change) and Figure 9 (50°–30° slope change) compare the simulation results and the experimental results for the presented dimensionless dynamic pressure distribution for different points on the stepped spillway. The results show a good agreement with the experimental and numerical simulations in all curves. For some points, few discrepancies can be noted in pressure magnitudes between the simulated and the observed ones, but they are in the acceptable range. Although the experimental data do not completely agree with the simulated results, there is an overall agreement.

Figure 8 Comparison between simulated and experimental results for the dimensionless pressure for steps number  −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.
Figure 8 Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

Figure 8VIEW LARGEDOWNLOAD SLIDE

Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 +3 and +20 on the horizontal step faces of 50°–18.6° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

Figure 9 Comparison between simulated and experimental results for the dimensionless pressure for steps number  −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.
Figure 9 Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

Figure 9VIEW LARGEDOWNLOAD SLIDE

Comparison between simulated and experimental results for the dimensionless pressure for steps number −1, −2, −3 and +1, +2 and +30, +31 on the horizontal step face of 50°–30° slope change configuration, for dc/h = 4.6, x is the distance from the step edge.

The pressure profiles were acquired at an acquisition time equal to 70 seconds in skimming flow on 50°–18.6°, where p is the measured dynamic pressure, h is step height and ϒ is water specific weight. A negative sign for steps represents a step upstream of the slope change cross-section and a positive sign represents a step downstream of the slope change cross-section.

Figure 10 shows the experimental streamwise development of dimensionless pressure on the 50°–18.6° slope change for dc/h = 4.6, x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute compared with the numerical simulation. It is obvious from Figure 10 that the streamwise development of dimensionless pressure before slope change (steps number −1, −2 and −3) both of the experimental and simulated results are close to each other. However, it is clear that there is a little difference between the results of the streamwise development of dimensionless pressure at step numbers +1, +2 and +3. Moreover, from step number +3 to the end, the curves get close to each other.

Figure 10 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.
Figure 10 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.

Figure 10VIEW LARGEDOWNLOAD SLIDE

Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–18.6° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.3 on 18.6° sloping chute.

Figure 11 compares the experimental and the numerical results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute. It is apparent that the outcomes of the experimental work are close to the numerical results, however, the results of the simulation are above the experimental ones before the slope change, but the results of the simulation descend below the experimental ones after the slope change till the end.

Figure 11 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.
Figure 11 Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.

Figure 11VIEW LARGEDOWNLOAD SLIDE

Comparison between experimental and simulated results for the streamwise development of the dimensionless pressure on the 50°–30° slope change, for dc/h = 4.6, and x/l = 0.35 on 50° sloping chute and x/l = 0.17 on 30° sloping chute.

CONCLUSION

ListenReadSpeaker webReader: Listen

In this research, numerical modelling was attempted to investigate the effect of abrupt slope change on the flow properties (air entrainment, velocity distribution and dynamic pressure) over a stepped spillway with two different models and various flow rates in a skimming flow regime by using the CFD technique. The numerical model was verified and compared with the experimental results of Ostad Mirza (2016). The same domain of the numerical model was inputted as in experimental models to reduce errors as much as possible.

Flow-3D is a well modelled tool that deals with particles. In this research, the model deals well with air entrainment particles by observing their results with experimental results. And the reason for the small difference between the numerical and the experimental results is that the program deals with particles more accurately than the laboratory. In general, both numerical and experimental results showed that near to the slope change the flow bulking, air entrainment, velocity distribution and dynamic pressure are greatly affected by abrupt slope change on the steps. Although the extent of the slope change was relatively small, the influence of the slope change was major on flow characteristics.

The Renormalized Group (RNG) model was selected as a turbulence solver. For 3D modelling, orthogonal mesh was used as a computational domain and the mesh grid size used for X, Y, and Z direction was equal to 0.015 m. In CFD modelling, air concentration and velocity distribution were recorded for a period of 25 seconds, but dynamic pressure was recorded for a period of 70 seconds. The results showed that there is a good agreement between the numerical and the physical models. So, it can be concluded that the proposed CFD model is very suitable for use in simulating and analysing the design of hydraulic structures.

이 연구에서 수치 모델링은 두 가지 다른 모델과 다양한 유속을 사용하여 스키밍 흐름 영역에서 계단식 배수로에 대한 유동 특성 (공기 혼입, 속도 분포 및 동적 압력)에 대한 급격한 경사 변화의 영향을 조사하기 위해 시도되었습니다. CFD 기술. 수치 모델을 검증하여 Ostad Mirza (2016)의 실험 결과와 비교 하였다. 오차를 최대한 줄이기 위해 실험 모형과 동일한 수치 모형을 입력 하였다.

Flow-3D는 파티클을 다루는 잘 모델링 된 도구입니다. 이 연구에서 모델은 실험 결과를 통해 결과를 관찰하여 공기 혼입 입자를 잘 처리합니다. 그리고 수치와 실험 결과의 차이가 작은 이유는 프로그램이 실험실보다 입자를 더 정확하게 다루기 때문입니다. 일반적으로 수치 및 실험 결과는 경사에 가까워지면 유동 벌킹, 공기 혼입, 속도 분포 및 동적 압력이 계단의 급격한 경사 변화에 크게 영향을받는 것으로 나타났습니다. 사면 변화의 정도는 상대적으로 작았지만 사면 변화의 영향은 유동 특성에 큰 영향을 미쳤다.

Renormalized Group (RNG) 모델이 난류 솔버로 선택되었습니다. 3D 모델링의 경우 계산 영역으로 직교 메쉬가 사용되었으며 X, Y, Z 방향에 사용 된 메쉬 그리드 크기는 0.015m입니다. CFD 모델링에서 공기 농도와 속도 분포는 25 초 동안 기록되었지만 동적 압력은 70 초 동안 기록되었습니다. 결과는 수치 모델과 물리적 모델간에 좋은 일치가 있음을 보여줍니다. 따라서 제안 된 CFD 모델은 수력 구조물의 설계 시뮬레이션 및 해석에 매우 적합하다는 결론을 내릴 수 있습니다.

DATA AVAILABILITY STATEMENT

ListenReadSpeaker webReader: Listen

All relevant data are included in the paper or its Supplementary Information.

REFERENCES

Boes R. M. Hager W. H. 2003a Hydraulic design of stepped spillways. Journal of Hydraulic Engineering 129 (9), 671–679.
Google Scholar
Boes R. M. Hager W. H. 2003b Two-Phase flow characteristics of stepped spillways. Journal of Hydraulic Engineering 129 (9), 661–670.
Google Scholar
Chanson H. 1994 Hydraulics of skimming flows over stepped channels and spillways. Journal of Hydraulic Research 32 (3), 445–460.
Google Scholar
Chanson H. 1997 Air Bubble Entrainment in Free Surface Turbulent Shear Flows. Academic Press, London.
Google Scholar
Chanson H. 2002 The Hydraulics of Stepped Chutes and Spillways. Balkema, Lisse, The Netherlands.
Google Scholar
Felder S. Chanson H. 2011 Energy dissipation down a stepped spillway with nonuniform step heights. Journal of Hydraulic Engineering 137 (11), 1543–1548.
Google Scholar
Flow Science, Inc. 2012 FLOW-3D v10-1 User Manual. Flow Science, Inc., Santa Fe, CA.
Ghaderi A. Daneshfaraz R. Torabi M. Abraham J. Azamathulla H. M. 2020a Experimental investigation on effective scouring parameters downstream from stepped spillways. Water Supply 20 (5), 1988–1998.
Google Scholar
Ghaderi A. Abbasi S. Abraham J. Azamathulla H. M. 2020b Efficiency of trapezoidal labyrinth shaped stepped spillways. Flow Measurement and Instrumentation 72, 101711.
Google Scholar
Gonzalez C. A. Chanson H. 2008 Turbulence and cavity recirculation in air-water skimming flows on a stepped spillway. Journal of Hydraulic Research 46 (1), 65–72.
Google Scholar
Gunal M. 1996 Numerical and Experimental Investigation of Hydraulic Jumps. PhD Thesis, University of Manchester, Institute of Science and Technology, Manchester, UK.
Hirt C. W. Nichols B. D. 1981 Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics 39 (1), 201–225.
Google Scholar
Matos J. 2000 Hydraulic design of stepped spillways over RCC dams. In: Intl Workshop on Hydraulics of Stepped Spillways (H.-E. Minor & W. Hager, eds). Balkema Publ, Zurich, pp. 187–194.
Google Scholar
Mohammad Rezapour Tabari M. Tavakoli S. 2016 Effects of stepped spillway geometry on flow pattern and energy dissipation. Arabian Journal for Science & Engineering (Springer Science & Business Media BV) 41 (4), 1215–1224.
Google Scholar
Ostad Mirza M. J. 2016 Experimental Study on the Influence of Abrupt Slope Changes on Flow Characteristics Over Stepped Spillways. Communications du Laboratoire de Constructions Hydrauliques, No. 64 (A. J. Schleiss, ed.). Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland.
Roshan R. Azamathulla H. M. Marosi M. Sarkardeh H. Pahlavan H. Ab Ghani A. 2010 Hydraulics of stepped spillways with different numbers of steps. Dams and Reservoirs 20 (3), 131–136.
Google Scholar
Shahheydari H. Nodoshan E. J. Barati R. Moghadam M. A. 2015 Discharge coefficient and energy dissipation over stepped spillway under skimming flow regime. KSCE Journal of Civil Engineering 19 (4), 1174–1182.
Google Scholar
Takahashi M. Ohtsu I. 2012 Aerated flow characteristics of skimming flow over stepped chutes. Journal of Hydraulic Research 50 (4), 427–434.
Google Scholar
Versteeg H. K. Malalasekera W. 2007 An Introduction to Computational Fluid Dynamics: The Finite Volume Method. Pearson Education, Harlow.
Google Scholar
© 2021 The Authors
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).

Figure 1. Sketch map of the port Laozi on Lake Hongze

FLOW-3D software for substantiation the layout of the port water area

항구 수역의 레이아웃을 입증하기 위한 FLOW-3D 소프트웨어

B Pan1

  • and N Belyaev2
    1 Peter the Great St. Petersburg Polytechnic University, St. Petersburg, Russia
    pan3.b@edu.spbstu.ru

Abstract

방파제 설계시 항만 내 수역 및 접안 시설에서 중대한 파도 발생을 배제 할 필요가 있다. 이 기사는 항구 지역의 물 이동 계산 결과를 제시합니다. 항구 입구의 위치와 주어진 물체의 크기에 대한 가장 안전한 옵션이 확인되었습니다. FLOW – 3D 프로그램을 사용하여 항구 수역의 예비 배치 단계에서 계획되고 건설적인 솔루션을 선택할 가능성이 입증되었습니다.

Introduction

항구를 설계 할 때 계류선의 가장 합리적인 구성과 항구 수역의 레이아웃을 선택하여 항구의 영토를 배치하는 것이 필요합니다. 러시아 연방의 항구 수역 배치는 항구 수역 요소에 대한 사양을 포함하는 해로, 페어웨이 및 기동 구역에 대한 설계 표준의 요구 사항에 따라 수행됩니다 [1, 2].

항구 물은 파도, 퇴적물 축적 및 얼음으로부터 보호되어야 합니다. 항구 계획을 작성할 때, 선박의 기동 및 연안 계류 중뿐만 아니라 선적 및 하역 및 기타 항구 운영 중에 선박 계류에 대한 정상적인 조건이 생성되도록 파도로부터 수역을 보호하는 정도를 제공해야 합니다.

설계 결정은 새로운 포트를 설계하거나 기존 포트를 개발할 때 물리적 또는 수학적 모델링을 기반으로 합니다 [2]. 항구 수역에서 계산 된 물 흐름의 매개 변수는 수문 기상 조사, 장기 현장 관찰 및 실험실 연구의 결과를 기반으로 하도록 권장됩니다.

공학 수문 기상 측량 데이터가 불충분하면 계산 방법을 기반으로 설계 폭풍의 파도 매개 변수를 결정할 수 있습니다. 사용된 계산 방법이 국제 실무에서 동일한 목적으로 채택된 방법 (모델)에 부합하는지 표시하는 것이 좋습니다 [3].

Figure 1. Sketch map of the port Laozi on Lake Hongze
Figure 1. Sketch map of the port Laozi on Lake Hongze
Figure 2. The location of the port entrance on Lake Hongze: a – variant 1; b – variant 2; c – variants 3-5
Figure 2. The location of the port entrance on Lake Hongze: a – variant 1; b – variant 2; c – variants 3-5
Figure 3. Port water area plan
Figure 3. Port water area plan
Figure 4. Modeling of variant 1 with the movement of waves in the port water area
Figure 4. Modeling of variant 1 with the movement of waves in the port water area
Figure 5. Modeling of variant 2: a is prevailing movement of water towards the enclosed water area; b is prevailing reverse movement of water
Figure 5. Modeling of variant 2: a is prevailing movement of water towards the enclosed water area; b is prevailing reverse movement of water
Figure 6. Modeling of variant 3
Figure 6. Modeling of variant 3
Figure 7. Modeling of variant 4
Figure 7. Modeling of variant 4
Figure 8. Modeling of variant 5
Figure 8. Modeling of variant 5
Figure 9. Plan of the port water area with design points
Figure 9. Plan of the port water area with design points
Figure 10. Change in water depth at point A: a – variant 1; b – variant 2
Figure 10. Change in water depth at point A: a – variant 1; b – variant 2
Figure 11. Change in water depth at point A: a – variant 3; b – variant 4; c – variant 5
Figure 11. Change in water depth at point A: a – variant 3; b – variant 4; c – variant 5
Figure 12. Change in water depth at points A (a) and C (b) for variant 3
Figure 12. Change in water depth at points A (a) and C (b) for variant 3
Figure 13. Change in water depth at points A (a) and B (b) for variant 3
Figure 13. Change in water depth at points A (a) and B (b) for variant 3
Figure 14. Scheme of vessel traffic: a – variant 3; b – variant 4
Figure 14. Scheme of vessel traffic: a – variant 3; b – variant 4

References
[1] SP 350.1326000.2018. 2018 Norms for technological design of sea ports (Moscow:
Standartinform) p 226
[2] SP 444.1326000.2019. 2019 Standards for the design of sea channels, fairways and
maneuvering areas (Moscow: Standartinform) p 62
[3] SP 38.13330.2012. 2014 Loads and impacts on Hydraulic structures (from wave, ice and ships)
(Moscow: Ministry of Regional Development of the Russian Federation) p 112
[4] Rijnsdorp D P Smit P B and Zijlema M 2012 Non-hydrostatic modelling of infragravity waves
using SWASH. Proceedings of 33rd Conference on Coastal Engineering. pp 1287–1299
[5] Kantardgi I G Zheleznyak M J 2016 Laboratory and numerical study of waves in the port area.
Magazine of Civil Engineering No 6 pp 49–59 DOI: 10.5862/MCE.66.5
[6] Zheleznyak M J Kantardgi I G Sorokin M S and Polyakov A I 2015 Resonance properties of
seaport water areas Magazine of Civil Engineering № 5(57) pp 3-19 DOI:10.5862/MCE.57.1
[7] Kantarzhi I Zuev N Shunko N 2014 Numerical and physical modelling of the waves inside the
new marina in Gelendjik (Black Sea) Application of physical modelling to port and coastal
protection. Proceedings of 5th international conference Coastlab (Varna) Vol 2 pp 253–262
[8] Makarov K N and Chebotarev A G 2015 Breakwater placement at the root of a seawall
Magazine of Civil Engineering № 3(55) pp 67-78 DOI: 10.5862/MCE.55.8
[9] Belyaev N D Lebedev V V and Alexeeva A V 2017 Investigation of the soil structure changes
under the tsunami waves impact on the marine hydrotechnical structures V 10 № 4 pp 44-52
DOI: 10.7868/S2073667317040049
[10] Lebedev V V Nudner I S and Belyaev N D 2018 The formation of the seabed surface relief near
the gravitational object Magazine of Civil Engineering No 79(3) pp 120–131 DOI:
10.18720/MCE.79.13
[11] Kofoed-Hansen H Sloth P Sørensen O R Fuchs J 2000 Combined numerical and physical
modelling of seiching in exposed new marina Proceedings of 27th international conference of
coastal engineering pp 3600–3614
[12] Smit P Stelling G and Zijlema M 2011 Assessment of nonhydrostatic wave-flow model
SWASH for directionally spread waves propagating through a barred basin Proceedings of
ACOMEN 2011 pp 1–10
[13] Zijlema M Stelling G Smit P 2011 SWASH: An operational public domain code for simulating
wave fields and rapidly varied flows in coastal waters. Coastal Engineering. № 10(58). pp 992–
1012
[14] FLOW-3D® 2008 User’s Manual Version 9.3 Flow Science Inc p 821
[15] Pan Bayan and Belyaev N D 2019 Week of Science SPbPU: Proceedings of an international
scientific conference The best reports. pp 3-7
[16] Girgidov A A 2011 Hybrid simulation in hydrotechnical facilities design and FLOW-3D as a
tool its realization Magazine of Civil Engineering №3 pp 21-27
[17] Girgidov A A 2010 Proceeding of the VNIIG vol 260. pp 12-19

[18] Vasquez J A Walsh B W 2009 CFD simulation of local scour in complex piers under tidal flow,
33rd IAHR Congress: Water Engineering for a Sustainable Environment, © 2009 by
International Association of Hydraulic Engineering & Research (IAHR) ISBN: 978-94-90365-
01-1.
[19] Shan-Hwei Ou Tai-Wen Hsu and Jian-Feng Lin 2010 Experimental and Numerical Studies on
Wave Transformation over Artificial Reefs Proceedings of the International Conference on
Coastal Engineering (Shanghai, China) No 32
[20] Hirt C and Nichols B 1980 Volume of Fluid Method for the Dynamics of Free Boundaries
Journal Comp. Phys 39 p 201.

Capsule-type Vane Tank

Numerical Simulation Analysis of Liquid Transportation in Capsule-type Vane Tank under Microgravity

Microgravity 하에서 캡슐형 베인 탱크의 액체 수송에 대한 수치 시뮬레이션 분석

Abstract

Li Yong-Qiang1,2 & Dong Jun-Yan1 & Rui Wei
1Received: 17 June 2019 /Accepted: 4 December 2019
#Springer Nature B.V. 2020

In order to research the influence of the guide vane on liquid transmission performance in a tank under microgravity, simulation analysis was carried out with FLOW-3D software. Firstly, it compared the working condition under the charging rate of 10% with the corresponding experiment results of the drop tower and validated the correctness of the simulation process. And then it changed the structure parameters of the guide vane, researched the influence of different quantity, gap and thickness on climbing rate of liquid, and analyzed the causing reasons of these effects in-depth. This paper provided a reference for the design of internal guiding vane of microgravity tank.

본 논문은 가이드 베인이 미세 중력 상태의 탱크에서 액체 전달 성능에 미치는 영향을 연구하기 위해 FLOW-3D 소프트웨어를 사용하여 시뮬레이션 분석을 수행했습니다. 첫째, 10 % 충전율 하에서 작업 조건을 드롭 타워의 해당 실험 결과와 비교하여 시뮬레이션 프로세스의 정확성을 검증했다. 그리고 가이드 베인의 구조 매개 변수를 변경하고, 액체의 상승 속도에 대한 양, 간격 및 두께의 영향을 연구하고 이러한 영향의 원인을 심도있게 분석했습니다. 이 논문은 미세 중력 탱크의 내부 안내 날개 설계에 대한 참고 자료를 제공했습니다.

Capsule-type Vane Tank
Capsule-type Vane Tank
The relationship curve between the square of climbing height and time with a = 6 mm
The relationship curve between the square of climbing height and time with a = 6 mm
The relationship curve between the vane’s liquid transportation and time under different width a
The relationship curve between the vane’s liquid transportation and time under different width a
Figure 7. Formation of incident and reflected waves.

Investigate Impact Force of Dam-Break Flow against Structures by Both 2D and 3D Numerical Simulations

2D 및 3D 수치 시뮬레이션에 의한 댐 붕괴유동의 구조물 충격력 조사

1 Faculty of Water Resources Engineering, Thuyloi University, 175 Tay Son, Dong Da, Ha Noi 116705, Vietnam
2 Hydraulic Construction Institute, 3/95 Chua Boc, Dong Da, Ha Noi 116705, Vietnam
* Author to whom correspondence should be addressed.
Academic Editor: Costanza Aricò
Water 2021, 13(3), 344;

Abstract

본 논문의 목적은 일부 2D 및 3D 수치 모델이 침수 지역에 고립된 건물 또는 건물 배열이 있는 곳에서 홍수 파동을 시뮬레이션하는 능력을 조사하는 것이었습니다.

먼저, 제안된 2D 수치 모델은 구조화된 메시에서 2D 얕은 물 방정식(2D-SWEs)을 해결하기 위한 유한 볼륨 방법(FVM)을 기반으로 했습니다.

FDS (flux-difference splitting)은 정확한 질량 균형을 얻기 위해 사용되었고 Roe 체계는 Riemann 문제를 근사하기 위해 호출되었습니다.

둘째, 상업적으로 이용 가능한 3D CFD 소프트웨어 패키지가 선택되었으며, 여기에는 두 가지 난류 모델이 포함된 Flow 3D 모델이 포함되어 있습니다.

RNG(Renormalized Group) 및 LES(Large-eddy Simulation)를 사용하는 레이놀즈 평균 Navier-Stokes(RAN)입니다. 댐 붕괴 흐름으로 인한 장애물에 대한 충격력의 수치 결과는 3D 솔루션이 2D 솔루션보다 훨씬 낫다는 것을 보여주었습니다.

건물 배열에 작용하는 충격력의 3D 수치 힘 결과를 보유하고 있는 실험 데이터와 비교함으로써, 속도 유도력이 동적 힘에 미치는 영향은 Froude 숫자의 함수와 사고 파동의 수심 함수에 의해 정량화 되었습니다. 또한, 우리는 힘의 강도의 피크 값의 3D 계산 결과에 대한 초기 물 단계와 댐 붕괴 폭의 영향을 조사했습니다.

The aim of this paper was to investigate the ability of some 2D and 3D numerical models to simulate flood waves in the presence of an isolated building or building array in an inundated area. Firstly, the proposed 2D numerical model was based on the finite-volume method (FVM) to solve 2D shallow-water equations (2D-SWEs) on structured mesh. The flux-difference splitting method (FDS) was utilized to obtain an exact mass balance while the Roe scheme was invoked to approximate Riemann problems. Secondly, the 3D commercially available CFD software package was selected, which contained a Flow 3D model with two turbulent models: Reynolds-averaged Navier-Stokes (RANs) with a renormalized group (RNG) and a large-eddy simulation (LES). The numerical results of an impact force on an obstruction due to a dam-break flow showed that a 3D solution was much better than a 2D one. By comparing the 3D numerical force results of an impact force acting on building arrays with the existence experimental data, the influence of velocity-induced force on a dynamic force was quantified by a function of the Froude number and the water depth of the incident wave. Furthermore, we investigated the effect of the initial water stage and dam-break width on the 3D-computed results of the peak value of force intensity.

Keywords: dam-break wave2D numerical modelFlow 3D modelstructuresimpact force

Introduction

홍수 위험 분석에 따른 도시 계획은 최근에 큰 연구 과제였습니다.

건물 또는 건물 그룹에 대한 홍수 파동의 영향에 대한 연구는 하류 지역에 대한 조기 경고 또는 안전 의식 향상에 중요한 역할을 했습니다. 기본적으로 댐 파괴 흐름에 대한 연구는 실험 측정이나 수치 시뮬레이션을 통해 추정 할 수 있습니다 [1,2,3,4,5,6].

컴퓨터 처리 능력의 증가로 인해 불연속 흐름에 대한 수치 연구가 비용 효율적이되었습니다. 지난 10 년 동안 얕은 물 솔버는 정확성과 계산 능력면에서 크게 향상되었습니다. 침수 가능 지역의 수심 및 속도 프로파일과 같은 유체 역학적 매개 변수에 많은주의를 기울였습니다 [1,2,3,4,5,6,7,8].

Migot et al. [9]는 도시 홍수의 실험적 모델링에 관한 많은 기사를 검토했습니다. 그 논문에 언급 된 45 개의 작품 중 단 4 개의 프로젝트 만이 장애물에 가해지는 일정한 또는 비정상적인 흐름의 힘 또는 압력을 측정했습니다.

또한 물리적 및 2D 수치 모델에서 건물 또는 건물 그룹에 돌발 홍수가 미치는 영향에 대한 연구는 거의 없었습니다. 얕은 물 모델은 [10,11]에서 고립된 장애물에 대한 충격의 힘을 예측하는데 사용되었습니다.

한편 Shige-eda [12]는 액체와 건물 배열 간의 상호 작용을 결정하기 위해 물리적 모델과 2D 수치 체계를 선택했습니다. Aureli와 Shige-eda는 수직 속도와 가속도를 무시하기 때문에 댐 파괴 흐름의 힘을 추정하기 위한 2D 얕은 물 방정식 (SWE)의 단점을 보여주었습니다 [10,12].

Migot [9]은 또한 장애물 주변의 시뮬레이션된 홍수 흐름에 대한 2D SWE에 대한 여러 출판물이 있었지만 이 주제에 대한 3D 수치 모델에 대한 연구는 거의 없다고 지적했습니다. 최근 전산 유체 역학 (CFD) 3D 시뮬레이션은 유체 흐름과 관련된 문제를 해결하기위한 광범위한 도구가되었습니다.

댐 파괴 파의 특성은 [13,14,15,16]에 의해 주목되었고 Issakhov [17]는 다양한 종류의 장애물이 압력 분포에 미치는 영향을 조사하기 위해 CFD 방법을 사용했습니다. 그들은 분포가 댐 표면에서 3 배 더 낮다는 것을 밝혔다.

Aureli [10]는 댐 파괴 파가 구조물에 미치는 영향의 정적 힘을 평가하기 위해 실험 테스트와 2D 및 3D 수치 모델을 사용했습니다. Mokarani [18]는 댐 브레이크 흐름 영향의 VOF 시뮬레이션에서 피크 압력 안정성 조건을 연구했습니다.

앞서 언급한 작품에서 구조물이나 구조물 군에 작용하는 힘은 압력에 의한 정 수력 또는 정력이었다. 한편, 급류에서 속도로 인한 힘은 압력 력보다 크거나 같았습니다 [19]. Armanini [20]는 정상 흐름에 대해이 항을 추정하기 위한 분석적 표현 만을 제시했습니다. 우리가 아는 한, 건물 그룹에 작용하는 비정상 흐름의 동적 힘을 생성하기 위해 2D 및 3D 수학적 모델을 모두 사용하는 작업은 없습니다.

따라서 본 연구에서는 제안된 2D 수치 모델과 3D 수학적 모델 모두에 의해 고립 된 장애물 또는 장애물 그룹에 대한 급격한 비정상 흐름의 테스트 사례를 재현했습니다. 수심 및 유속 수문 그래프와 같은 몇 가지 수력 학적 특성이 추정되었으며 측정 된 데이터와 매우 잘 일치했습니다.

특히 댐 브레이크 흐름이 서로 다른 건물에 가하는 동적인 힘도 시뮬레이션했습니다. 속도 유도 힘이 동적 힘에 미치는 영향 수준을 나타내는 매개 변수는 Froude 수와 입사 파동의 수심의 함수인 것으로 밝혀졌습니다. 또한 붕괴된 댐 사이트 폭 (b)과 초기 수위 (h0)는 충격력의 최대 값에 영향을 미치는 변수로 고려되었습니다.

Figure 1. (a) Configuration of experiment test (dimension in meters); (b) Gauges on the vertical front face of building.
Figure 1. (a) Configuration of experiment test (dimension in meters); (b) Gauges on the vertical front face of building.
Figure 2. (a) Distributed pressure profiles at centerline of front face of column; (b) Comparison of load-time histories simulated by different numerical models
Figure 2. (a) Distributed pressure profiles at centerline of front face of column; (b) Comparison of load-time histories simulated by different numerical models
Figure 3. Group of buildings in flooded area.
Figure 3. Group of buildings in flooded area.
Figure 4. Water depth and u-velocity profiles at gauge b.
Figure 4. Water depth and u-velocity profiles at gauge b.
Figure 5. Water hydrographs at gauges a and c.
Figure 5. Water hydrographs at gauges a and c.
Figure 6. Velocity component profiles at gauges a and c.
Figure 6. Velocity component profiles at gauges a and c.
Figure 7. Formation of incident and reflected waves.
Figure 7. Formation of incident and reflected waves.
Figure 8. Snapshots of streamlines of Froude number at different times: 1.0 s, 2.0 s, 5.0 s and 10 s.
Figure 8. Snapshots of streamlines of Froude number at different times: 1.0 s, 2.0 s, 5.0 s and 10 s.
Figure 9. Force in the flow direction exerted on 6 buildings.
Figure 9. Force in the flow direction exerted on 6 buildings.
Figure 10. The linear regression between forces per unit width (F) and q2b/h0.
Figure 10. The linear regression between forces per unit width (F) and q2b/h0.

Conclusions

댐 붕괴 흐름으로 인한 홍수 파도는 높은 속도 또는 큰 깊이가 관련되었을 때 건물에 큰 영향을 미칩니다. 본 논문에서는 2D 및 3D 수치 모델의 건물 및 건물 그룹에 대한 빠른 흐름에 의해 발생하는 유압 특성과 충격 부하를 추정할 수 있는 능력을 조사했습니다. 얕은 물 방정식에 기초한 2D 수학 모델은 FDS 방법으로 해결되었으며, FDS 방법은 최신 버전의 Flow 3D 유체 역학 모델과 함께 사용되었습니다. 연구의 주요 발견은 다음과 같습니다.
(1) 수심 또는 속도 프로파일을 공식화하기 위해 2D 및 3D 수치 솔루션은 모두 매우 유사합니다. 제안된 2D 수치 모델은 정적 힘의 최대 값 뿐만 아니라 수심 및 속도 구성 요소를 포함하는 유압 특성을 예측하는 데 적합합니다. 그러나 LES 및 RAN 난류 모듈이 포함된 3D 유체역학 모델은 2D 얕은 흐름 모델이 1개만 제공하는 동안 두 개의 최고 충격 부하를 잘 포착할 수 있습니다. 일반적으로 3D 결과는 실험 결과와 더 가깝습니다.
(2) 여러 건물에 대한 정적 및 동적 힘은 모두 LES 모듈을 사용하여 Flow 3D에 의해 계산되었습니다. 건물에서 속도에 의한 힘과 압력의 역할은 위치에 따라 다릅니다. 댐 현장 근처에서, 속도 유도 힘은 댐 파괴 파동의 주 방향에서 멀리 떨어져 있거나 두 번째 배열에서 압력 힘이 더 중요합니다. 속도 유도 힘의 영향은 매개 변수 α에 의해 정량화되며, 이는 사고파의 Froude 숫자와 수심 함수로 수행됩니다. q2b/h0과 정적 힘과 동적 힘의 피크 강도 사이의 선형 회귀 관계는 합리적인 R-제곱 양으로 해결됩니다.

추가 연구에서, 제시된 2D 수치 모델의 견고성과 효과는 더 명확하게 드러날 것입니다. 대규모 도메인에 대한 홍수 흐름을 시뮬레이션하는 데 쉽게 적용할 수 있습니다. 또한, α 매개변수의 제안된 방정식(21)은 실제 사례 연구에서 다운스트림 영역의 건물에 대한 속도 유도 힘의 영향을 정확하게 평가하기 위한 매우 의미가 있습니다. 이 매개 변수의 정확도 수준을 높이려면 서로 다른 조건에서 장애물에 작용하는 여러 가지 힘 실험이 구현되어야 합니다.

References

  1. Testa, G.; Zuccala, D.; Alcrudo, F.; Mulet, J.; Frazao, S.S. Flash flood flow experiment in a simplifed urban district. J. Hydraul. Res. 200745, 37–44. [Google Scholar] [CrossRef]
  2. Soares-Frazao, S.; Zech, Y. Dam-break flow through an idealized city. J. Hydraul. Res. 200846, 648–665. [Google Scholar] [CrossRef]
  3. Soares-Frazão, S.; Zech, Y. Experimental study of dam-break flow against an isolated obstacle. J. Hydraul. Res. 200745, 27–36. [Google Scholar] [CrossRef]
  4. Soares-Frazão, S. Experiments of dam-break wave over a triangular bottom sill. J. Hydraul. Res. 200745, 19–26. [Google Scholar] [CrossRef]
  5. di Cristo, C.; Evangelista, S.; Greco, M.; Iervolino, M.; Leopardi, A.; Vacca, A. Dam-break waves over an erodible embankment: Experiments and simulations. J. Hydraul. Res. 201856, 196–210. [Google Scholar] [CrossRef]
  6. Evangelista, S. Experiments and numerical simulations of dike erosion due to a wave impact. Water 20157, 5831–5848. [Google Scholar] [CrossRef]
  7. Li, Y.L.; Yu, C.H. Research on dam break flow induced front wave impacting a vertical wall based on the CLSVOF and level set methods. Ocean Eng. 2019178, 442–462. [Google Scholar] [CrossRef]
  8. Özgen, I.; Zhao, J.; Liang, D.; Hinkelmann, R. Urban flood modeling using shallow water equations with depth-dependent anisotropic porosity. J. Hydrol. 2016541, 1165–1184. [Google Scholar] [CrossRef]
  9. Mignot, E.; Li, X.; Dewals, B. Experimental modelling of urban flooding: A review. J. Hydrol. 2019568, 334–342. [Google Scholar] [CrossRef]
  10. Aureli, F.; Dazzi, A.; Maranzoni, A.; Mignosa, P.; Vacondio, R. Experimental and numerical evaluation of the force due to the impact of a dam break wave on a structure. Adv. Water Resour. 201576, 29–42. [Google Scholar] [CrossRef]
  11. Milanesi, L.; Pilotti, M.; Belleri, A.; Marini, A.; Fuchs, S. Vulnerability to flash floods: A simplified structural model for masonry buldings. Water Resour. Res. 201854, 7177–7197. [Google Scholar] [CrossRef]
  12. Shige-eda, M.; Akiyama, J. Numerical and experimental study on two dimensional flood flows with and without structures. J. Hydraul. Eng. 2003129, 817–821. [Google Scholar] [CrossRef]
  13. Cagatay, H.O.; Kocaman, S. Dam break flows during initial stage using SWE and RANs approaches. J. Hydraul. Res. 201048, 603–611. [Google Scholar] [CrossRef]
  14. Yang, S.; Yang, W.; Qin, S.; Li, Q.; Yang, B. Numerical study on characteristics of dam break wave. Ocean Eng. 2018159, 358–371. [Google Scholar] [CrossRef]
  15. Robb., D.M.; Vasquez., J.A. Numerical simulation of dam break flows using depth averaged hydrodynamic and three dimensional CFD models. In Proceedings of the 22nd Canadian Hydrotechnical Conference, Ottawa, ON, Canada, 28–30 April 2015. [Google Scholar]
  16. Kocaman, S.; Evangelista, S.; Viccione, G.; Guzel, H. Experimental and Numerical analysis of 3D dam break waves in an enclosed domain with a single oriented obstacles. Environ. Sci. Proc. 20202, 35. [Google Scholar] [CrossRef]
  17. Issakhov, A.; Zhandaulet, Y.; Nogaeva, A. Numerical simulation of dambreak flow for various forms of the obstacle by VOF method. Int. J. Multiph. Flow 2018109, 191–206. [Google Scholar] [CrossRef]
  18. Mokarani, C.; Abadie, S. Conditions for peak pressure stability in VOF simulations of dam break flow impact. J. Fluids Struct. 201662, 86–103. [Google Scholar] [CrossRef]
  19. Liu, L.; Sun, J.; Lin, B.; Lu, L. Building performance in dam break flow—an experimental sudy. Urban Water J. 201815, 251–258. [Google Scholar] [CrossRef]
  20. Armanini, A.; Larcher, M.; Odorizzi, M. Dynamic impact of a debris flow front against a vertical wall. In Proceedings of the 5th international conference on debris-flow hazards mitigation: Mechanics, prediction and assessment, Padua, Italy, 14–17 June 2011. [Google Scholar] [CrossRef]
  21. Hubbard, M.E.; Garcia Navarro, P. Flux difference splitting and the balancing of source terms and flux gradients. J. Comput. Phys. 2000165, 89–125. [Google Scholar] [CrossRef]
  22. Roe, P.L. A basis for upwind differencing of the two-dimensional unsteady Euler equations. In Numerical Methods in Fluids Dynamics II; Oxford University Press: Oxford, UK, 1986. [Google Scholar]
  23. Bradford, S.F.; Sander, B. Finite volume model for shallow water flooding of arbitrary topography. J. Hydraul. Eng. (ASCE) 2002128, 289–298. [Google Scholar] [CrossRef]
  24. Brufau, P.; Garica-Navarro, P. Two dimensional dam break flow simulation. Int. J. Numer. Meth. Fluids 200033, 35–57. [Google Scholar] [CrossRef]
  25. Hien, L.T.T. 2D Numerical Modeling of Dam-Break Flows with Application to Case Studies in Vietnam. Ph.D. Thesis, Brescia University, Brescia, Italy, 2014. [Google Scholar]
  26. Hien, L.T.T.; Tomirotti, M. Numerical modeling of dam break flows over complex topography. Case studies in Vietnam. In Proceedings of the 19th IAHR-APD Congress 2014, Hanoi, Vietnam, 21–24 September 2014; ISBN 978-604-82-1383-1. [Google Scholar]
  27. Flow-3D, Version 12.0; User Mannual; Flow Science Inc.: Santa Fe, NM, USA, 2020.
  28. Guney, M.S.; Tayfur, G.; Bombar, G.; Elci, S. Distorted physical model to study sudden partial dam break flow in an urban area. J. Hydraul. Eng. 2014140, 05014006. [Google Scholar] [CrossRef]
  29. Shige-eda, M.; Akiyama, J. Discussion and Closure to “Numerical and experimental study on two dimensional flood flows with and without structures” by Mirei Shige-eda and Juichiro Akiyama. J. Hydraul. Eng. 2005131, 336–337. [Google Scholar] [CrossRef]
  30. Ritter, A. Die Fortpflanzung der Wasserwelle (Generation of the water wave). Z. Ver. Dtsch. Ing. 189236, 947–954. [Google Scholar]
Fig. 9 (a) Velocity field, keyhole profile, and breakage of the keyhole to form bubble and (b) 2D temperature and velocity field along the longitudinal section

A Numerical Study on the Keyhole Formation During Laser Powder Bed Fusion Process

Keyhole에 대한 수치적 연구 : 레이저 분말 중 형성 베드 퓨전 공정

Subin Shrestha1
J.B. Speed School of Engineering,University of Louisville,Louisville, KY 40292
e-mail: subin.shrestha@louisville.edu

Y. Kevin Chou
J.B. Speed School of Engineering,University of Louisville,Louisville, KY 40292
e-mail: kevin.chou@louisville.edu

LPBF (Laser Powder Bed fusion) 공정 중 용융 풀의 동적 현상은 복잡하고 공정 매개 변수에 민감합니다. 에너지 밀도 입력이 특정 임계 값을 초과하면 키홀이라고 하는 거대한 증기 함몰이 형성 될 수 있습니다.

이 연구는 수치 분석을 통해 LPBF 과정에서 키홀 거동 및 관련 기공 형성을 이해하는 데 중점을 둡니다. 이를 위해 이산 분말 입자가 있는 열 유동 모델이 개발되었습니다.

이산 요소 방법 (DEM)에서 얻은 분말 분포는 계산 영역에 통합되어 FLOW-3D를 사용하는 3D 프로세스 물리학 모델을 개발합니다.

전도 모드 중 용융 풀 형성과 용융의 키홀 모드가 식별되고 설명되었습니다. 높은 에너지 밀도는 증기 기둥의 형성으로 이어지고 결과적으로 레이저 스캔 트랙 아래에 구멍이 생깁니다.

또한 다양한 레이저 출력과 스캔 속도로 인한 Keyhole 모양을 조사합니다. 수치 결과는 동일한 에너지 밀도에서도 레이저 출력이 증가함에 따라 Keyhole크기가 증가 함을 나타냅니다. Keyhole은 더 높은 출력에서 ​​안정되어 레이저 스캔 중 Keyhole 발생을 줄일 수 있습니다.

The dynamic phenomenon of a melt pool during the laser powder bed fusion (LPBF) process is complex and sensitive to process parameters. As the energy density input exceeds a certain threshold, a huge vapor depression may form, known as the keyhole. This study focuses on understanding the keyhole behavior and related pore formation during the LPBF process through numerical analysis. For this purpose, a thermo-fluid model with discrete powder particles is developed. The powder distribution, obtained from a discrete element method (DEM), is incorporated into the computational domain to develop a 3D process physics model using flow-3d. The melt pool formation during the conduction mode and the keyhole mode of melting has been discerned and explained. The high energy density leads to the formation of a vapor column and consequently pores under the laser scan track. Further, the keyhole shape resulted from different laser powers and scan speeds is investigated. The numerical results indicated that the keyhole size increases with the increase in the laser power even with the same energy density. The keyhole becomes stable at a higher power, which may reduce the occurrence of pores during laser scanning.

Keywords: additive manufacturing, keyhole, laser powder bed fusion, porosity

Fig. 1 (a) Powder added to the dispenser platform and (b) powder particles settled over build plate after the recoating process
Fig. 1 (a) Powder added to the dispenser platform and (b) powder particles settled over build plate after the recoating process
Fig. 2 3D computational domain used for single-track simulation
Fig. 2 3D computational domain used for single-track simulation
Fig. 3 Temperature-dependent material properties of Ti-6Al-4V
Fig. 3 Temperature-dependent material properties of Ti-6Al-4V
Fig. 4 Powder and substrate melting during laser application
Fig. 4 Powder and substrate melting during laser application
Fig. 5 Melt region formed after complete melting and solidification
Fig. 5 Melt region formed after complete melting and solidification
Fig. 6 Melt pool boundary comparison between the experiment [25] and the simulation
Fig. 6 Melt pool boundary comparison between the experiment [25] and the simulation
Fig. 7 Equilibrium points during the formation of vapor column [27]
Fig. 7 Equilibrium points during the formation of vapor column [27]
Fig. 8 Multiple reflection vectors from the keyhole wall
Fig. 8 Multiple reflection vectors from the keyhole wall
Fig. 9 (a) Velocity field, keyhole profile, and breakage of the keyhole to form bubble and (b) 2D temperature and velocity field along the longitudinal section
Fig. 9 (a) Velocity field, keyhole profile, and breakage of the keyhole to form bubble and (b) 2D temperature and velocity field along the longitudinal section
Fig. 10 Fluid flow in the transverse direction during keyhole melting
Fig. 10 Fluid flow in the transverse direction during keyhole melting
Fig. 11 Melt pool boundary compared with the experiment [21] for 195 W laser power and 400 mm/s scan speed
Fig. 11 Melt pool boundary compared with the experiment [21] for 195 W laser power and 400 mm/s scan speed
Fig. 12 Melt region formed after complete melting and solidification
Fig. 12 Melt region formed after complete melting and solidification
Fig. 13 2D images of the pores formed at the beginning of the single track and their 3D-rendered morphology
Fig. 13 2D images of the pores formed at the beginning of the single track and their 3D-rendered morphology
Fig. 14 Pore number and volume from a different level of power with LED = 0.4 J/mm [29]
Fig. 14 Pore number and volume from a different level of power with LED = 0.4 J/mm [29]
Fig. 15 Keyhole shape at different time steps from different parameters: (a) P = 100 W, v = 250 mm/s, (b) P = 200 W, v = 500 mm/s, (c) P = 300 W, v = 750 mm/s, and (d) P = 400 W, v = 1000 mm/s
Fig. 15 Keyhole shape at different time steps from different parameters: (a) P = 100 W, v = 250 mm/s, (b) P = 200 W, v = 500 mm/s, (c) P = 300 W, v = 750 mm/s, and (d) P = 400 W, v = 1000 mm/s
Fig. 16 Intensity dependence in the relationship between vapor column and evaporation pressure [27]
Fig. 16 Intensity dependence in the relationship between vapor column and evaporation pressure [27]
Fig. 17 Temperature distribution when laser has moved 0.8 mm with P = 300 W, v = 750 mm/s and P = 400 W, v = 1000 mm/s
Fig. 17 Temperature distribution when laser has moved 0.8 mm with P = 300 W, v = 750 mm/s and P = 400 W, v = 1000 mm/s
Fig. 18 Melt region with different level of power with LED of 0.4 J/mm
Fig. 18 Melt region with different level of power with LED of 0.4 J/mm

References

[1] Bauereiß, A., Scharowsky, T., and Körner, C., 2014, “Defect Generation and
Propagation Mechanism During Additive Manufacturing by Selective Beam
Melting,” J. Mater. Process. Technol., 214(11), pp. 2522–2528.
[2] Gong, H., Rafi, K., Gu, H., Starr, T., and Stucker, B., 2014, “Analysis of Defect
Generation in Ti–6Al–4V Parts Made Using Powder Bed Fusion Additive
Manufacturing Processes,” Add. Manuf., 1(2014), pp. 87–98.
[3] Wang, Y., Kamath, C., Voisin, T., and Li, Z., 2018, “A Processing Diagram for
High-Density Ti-6Al-4V by Selective Laser Melting,” Rapid Prototyping J., 24
(9), pp. 1469–1478.
[4] Khairallah, S. A., and Anderson, A., 2014, “Mesoscopic Simulation Model of
Selective Laser Melting of Stainless Steel Powder,” J. Mater. Process. Technol.,
214(11), pp. 2627–2636.
[5] Yadroitsev, I., Gusarov, A., Yadroitsava, I., and Smurov, I., 2010, “Single Track
Formation in Selective Laser Melting of Metal Powders,” J. Mater. Process.
Technol., 210(12), pp. 1624–1631.
[6] Xia, M., Gu, D., Yu, G., Dai, D., Chen, H., and Shi, Q., 2016, “Influence of Hatch
Spacing on Heat and Mass Transfer, Thermodynamics and Laser Processability
During Additive Manufacturing of Inconel 718 Alloy,” Int. J. Mach. Tools
Manuf., 109(2016), pp. 147–157.
[7] Lee, Y., and Zhang, W., 2016, “Modeling of Heat Transfer, Fluid Flow and
Solidification Microstructure of Nickel-Base Superalloy Fabricated by Laser
Powder bed Fusion,” Add. Manuf., 12(2016), pp. 178–188.
[8] Wu, Y.-C., San, C.-H., Chang, C.-H., Lin, H.-J., Marwan, R., Baba, S., and
Hwang, W.-S., 2018, “Numerical Modeling of Melt-Pool Behavior in Selective
Laser Melting with Random Powder Distribution and Experimental
Validation,” J. Mater. Process. Technol., 254(2018), pp. 72–78.
[9] Khairallah, S. A., Anderson, A. T., Rubenchik, A., and King, W. E., 2016, “Laser
Powder-bed Fusion Additive Manufacturing: Physics of Complex Melt Flow and
Formation Mechanisms of Pores, Spatter, and Denudation Zones,” Acta
Materialia, 108(2016), pp. 36–45.
[10] Tan, J., Tang, C., and Wong, C., 2018, “A Computational Study on Porosity
Evolution in Parts Produced by Selective Laser Melting,” Metall. Mater. Trans.
A, 49A(8), pp. 3663–3673.
[11] Leitz, K.-H., Singer, P., Plankensteiner, A., Tabernig, B., Kestler, H., and Sigl,
L. J. M. P. R., 2017, “Multi-Physical Simulation of Selective Laser Melting,”
Metal Powder Report, 72(5), pp. 331–338.
[12] Zhao, C., Fezzaa, K., Cunningham, R. W., Wen, H., Carlo, F., Chen, L., Rollett,
A. D., and Sun, T., 2017, “Real-time Monitoring of Laser Powder Bed Fusion
Process Using High-Speed X-ray Imaging and Diffraction,” Sci. Rep., 7(1),
p. 3602.
[13] Parab, N. D., Zhao, C., Cunningham, R., Escano, L. I., Fezzaa, K., Everhart, W.,
Rollett, A. D., Chen, L., and Sun, T., 2018, “Ultrafast X-ray Imaging of Laser–
Metal Additive Manufacturing Processes,” J. Synchrotron Radiat., 25(5),
pp. 1467–1477.
[14] Cunningham, R., Zhao, C., Parab, N., Kantzos, C., Pauza, J., Fezzaa, K., Sun, T.,
and Rollett, A. D., 2019, “Keyhole Threshold and Morphology in Laser Melting
Revealed by Ultrahigh-Speed X-Ray Imaging,” Science, 363(6429), pp. 849–852.
[15] Shrestha, S., Starr, T., and Chou, K., 2019, “A Study of Keyhole Porosity in
Selective Laser Melting: Single Track Scanning With Micro-CT Analysis,”
ASME J. Manuf. Sci. Eng., 141(7), pp. 1–23.
[16] Ye, J., Rubenchik, A. M., Crumb, M. F., Guss, G., and Matthews, M. J., 2018,
“Laser Absorption and Scaling Behavior in Powder Bed Fusion Additive
Manufacturing of Metals,” Proceedings of the CLEO: Science and Innovations,
Optical Society of America, San Jose, CA, May 13–18, Optical Society of
America, p. JW2A.117.
[17] Mishra, B., and Rajamani, R. K., 1992, “The Discrete Element Method for the
Simulation of Ball Mills,” Appl. Math. Modell., 16(11), pp. 598–604.
[18] Yan, W., Qian, Y., Ge, W., Lin, S., Liu, W. K., Lin, F., and Wagner, G. J., 2018,
“Meso-Scale Modeling of Multiple-Layer Fabrication Process in Selective
Electron Beam Melting: Inter-Layer/Track Voids Formation,” Materials and
Design, 141(2018), pp. 210–219.
[19] Kloss, C., Goniva, C., Hager, A., Amberger, S., and Pirker, S., 2012, “Models,
Algorithms and Validation for Opensource DEM and CFD–DEM,” Prog.
Comput. Fluid Dynam. Int. J., 12(2–3), pp. 140–152.
[20] Escano, L. I., Parab, N. D., Xiong, L., Guo, Q., Zhao, C., Fezzaa, K., Everhart,
W., Sun, T., and Chen, L., 2018, “Revealing Particle-Scale Powder Spreading
Dynamics in Powder-Bed-Based Additive Manufacturing Process by
High-Speed X-Ray Imaging,” Sci. Rep., 8(1), p. 15079.
[21] Gong, H., Gu, H., Zeng, K., Dilip, J., Pal, D., Stucker, B., Christiansen, D., Beuth,
J., and Lewandowski, J. J., 2014, “Melt Pool Characterization for Selective Laser
Melting of Ti-6Al-4V Pre-Alloyed Powder,” Proceedings of the Solid Freeform
Fabrication Symposium, Austin, TX, Aug. 4–6, pp. 256–267.
[22] Mills, K. C., 2002, Recommended Values of Thermophysical Properties for
Selected Commercial Alloys, Woodhead Publishing, Cambridge, UK.
[23] Shrestha, S., and Chou, K., 2017, “A Build Surface Study of Powder-Bed
Electron Beam Additive Manufacturing by 3D Thermo-Fluid Simulation and
White-Light Interferometry,” Int. J. Mach. Tools Manuf., 121(2017), pp. 37–49.
[24] Cho, J.-H., and Na, S.-J., 2006, “Implementation of Real-Time Multiple
Reflection and Fresnel Absorption of Laser Beam in Keyhole,” J. Phys. D:
Appl. Phys., 39(24), p. 5372.
[25] Dilip, J., Zhang, S., Teng, C., Zeng, K., Robinson, C., Pal, D., and Stucker, B.,
2017, “Influence of Processing Parameters on the Evolution of Melt Pool,
Porosity, and Microstructures in Ti-6Al-4V Alloy Parts Fabricated by Selective
Laser Melting,” Prog. Add. Manuf., 2(3), pp. 157–167.
[26] Bertoli, U. S., Wolfer, A. J., Matthews, M. J., Delplanque, J.-P. R., and Schoenung,
J. M., 2017, “On the Limitations of Volumetric Energy Density as a Design
Parameter for Selective Laser Melting,” Mater. Des., 113(2017), pp. 331–340.
[27] Kroos, J., Gratzke, U., and Simon, G., 1993, “Towards a Self-Consistent Model of
the Keyhole in Penetration Laser Beam Welding,” J. Phys. D: Appl. Phys., 26(3),
p. 474.
[28] Martin, A., Calta, N., Hammons, J., Khairallah, S., Nielsen, M., Shuttlesworth, R.,
Sinclair, N., Matthews, M., Jeffries, J., and Willey, T., 2019, “Ultrafast Dynamics
of Laser-Metal Interactions in Additive Manufacturing Alloys Captured by In Situ
X-Ray Imaging,” Mater. Today Adv., 1(2019), p. 100002.
[29] Shrestha, S., Starr, T., and Chou, K., 2018, “Individual and coupled contributions
of laser power and scanning speed towards process-induced porosity in selective
laser melting,” Proceedings of the Solid Freeform Fabrication Symposium,
Austin, TX, Aug. 13–15, pp. 1400–1409.
[30] Hann, D., Iammi, J., and Folkes, J., 2011, “A Simple Methodology for Predicting
Laser-Weld Properties From Material and Laser Parameters,” J. Phys. D: Appl.
Phys., 44(44), p. 445401.
[31] Trapp, J., Rubenchik, A. M., Guss, G., and Matthews, M. J., 2017, “In Situ
Absorptivity Measurements of Metallic Powders During Laser Powder-bed
Fusion Additive Manufacturing,” Appl. Mat. Today, 9(2017), pp. 341–349.

Cad2Stl

FLOW-3D 유틸리티 프로그램 안내

이 문서에서는 FLOW-3D에서 사용할 수 있는 일부 Utility Program에 대해 설명합니다. 유틸리티 프로그램의 목적은 시뮬레이션을 수행할 때 반드시 필요한 것은 아니지만 특정 작업을 쉽게 수행할 수 있도록 돕는 것입니다. 각 개별 유틸리티의 사용법은 다음과 같습니다.

  1. 파일 변환 및 STL 품질 검사 도구

FLOW-3D는 중립 형식인 STL파일 형식만 지원하며 대부분의 CAD 패키지에서 STL형식을 지원하지만 형상을 STL형식으로 만들 수 없는 이유가 있을 수 있습니다. 이로 인해 FLOW-3D 사용자는 여러 파일 변환 유틸리티를 사용할 필요가 있을 수 있습니다. 또한 STL 파일 품질을 확인하는데 사용할 수 있는 여러 유틸리티도 사용할 수 있습니다. 아래 나열된 이러한 유틸리티는 다음 섹션에서 자세히 설명합니다.

  • Cad2Stl : 다양한 CAD 형식에서 변환 파일을 사용하는.STL파일
  • Topo2STL : 파일을topo형식에서.STL파일로 변환하는 데 사용
  • MiniMagics :.STL파일의 오류를 확인하는 데 사용
  • qAdmesh :.STL파일의 오류를 확인하고 사소한 문제를 해결하는데 사용

Cad2Stl

Cad2Stl 은 다른 CAD 파일 형식을 FLOW-3D에서 사용되는 STL 파일 형식으로 변환하기 위한 파일 변환 도구입니다. Cad2Stl 은 다음 파일 형식을 STL 형식으로 변환합니다.

  • Autodesk 3D Max :.3ds
  • Autodesk 별명 :.obj
  • IGES: .igs,.iges
  • BREP :.brep
  • 단계 : .stp,.step
  • 아바쿠스 6.2+ :.inp
  • NASTRAN :.blk
  • Marc Mentat : 고정 형식과 쉼표로 구분.dat

Cad2Stl 은 파일에서 역 법선 벡터를 보정하는 기능도 있습니다. 이 유틸리티는 유지 보수 계약이 유효한 모든 FLOW-3D 고객에게 무료로 제공되며 FLOW-3D Usre Site의 유틸리티 페이지에서 다운로드 할 수 있습니다.

Cad2Stl 은 Flow Science Japan에서 FLOW-3D 사용자를 위해 개발되었습니다 .

Cad2Stl Program
  1. 변환 목록에 변환할 파일 추가
    • 추가 -변환 목록에 파일을 추가합니다.
    • 제거 -변환 목록에서 파일을 제거합니다. 제거하려면 변환 목록에서 파일을 강조 표시하고 제거를 선택하십시오.
    • 기본적으로 파일 이름은 import file 이름과 일치하는 CAD파일을 STL파일 이름으로 지정하는데 변경이 필요하면 더블 클릭하고 이름을 바꾸면 변경할 수 있습니다.
  2. 구체화 옵션을 사용하여 STL 파일의 품질을 선택하십시오. 선택하고 볼 수 있는 네 가지 수준의 정확도가 있습니다. 파일이 변환될 때마다 STL로 작성된 파일이 표시되므로 사용자가 만족스럽거나 더 높은 수준의 세분화가 필요한지 여부를 결정할 수 있습니다. 정확성이 향상되면 파일 크기는 증가하지만 처리 시간은 크게 증가하지 않습니다. 다른 파일 형식을 한 번에 로드하고 변환할 수 있습니다. 또한 변환 프로세스가 완료되면 파일을 로드하고 표시하기 위한 대화 상자가 열립니다. 이것은 BREP, IGES및 STEP 파일 형식에만 적용됩니다.
  3. 원하는 작업을 선택하십시오. 다른 파일 형식을 한 번에 로드하고 변환할 수 있습니다. 또한 변환 프로세스가 완료되면 파일을 로드하고 표시하기 위한 대화 상자가 열립니다.
    • 변환 -파일을 변환합니다. 한 파일을 변환하려면 로드할 파일 목록에서 해당 파일을 강조 표시하여 변환하십시오.
    • 모두 변환 -모든 파일을 변환
    • 표시 -변환된 파일을 강조 표시합니다
    • 면 방향 수정 -일반 수정 루틴
    • 변환 목록 숨기기 -더 나은 부품 표시를 위해 보기 화면을 증가 시킵니다.
    • 와이어 프레임 오버레이 -각 STL 패싯의 패싯 모서리를 오버레이 합니다. 이것은 오른쪽 하단의 확인란입니다.
    • 로그 지우기 – 변환 로그 텍스트 상자에 대한 모든 데이터 출력을 지웁니다.
  4. 종료 -프로그램을 닫습니다

qAdmesh

qAdmesh는 .STL파일에 오류 가 있는지 확인하는 도구이며 연결이 끊어진 패싯, 반전된 법선, 연결이 끊어진 패싯 및 누락된 패싯과 같은 사소한 문제를 해결하는 데 사용할 수 있습니다. qAdmesh를 시작하려면:

  • GUI에서: Model Setup 탭의 Tools ‣ qAdmesh로 이동하십시오.
  • Windows: 바탕 화면 아이콘을 클릭하거나 시작 메뉴에서 FLOW-3D v12.0 폴더의 형상 도구 하위 디렉토리에 있는 Admesh 항목으로 이동하십시오.
  • Linux의 경우: $F3D_HOME/utilities/qAdmesh을 실행하십시오.

명령: qAdmesh를 열고 찾아보기 버튼을 사용하여 지오메트리 파일을 로드 하십시오. 문제를 해결하고 수정 사항으로 새 형상 파일을 생성하려면 기본 옵션을 그대로 두고 출력 유형을 선택하고 새 형상 파일의 경로를 지정하십시오. 이진 STL 은 ASCII STL 옵션 보다 작은 파일을 생성하므로 권장됩니다 (이진 및 ASCII 형식 만 FLOW-3D 로 인식됨). 그런 다음 적용을 클릭하여 파일을 확인하고 수정하십시오.

qAdmesh program
qAdmesh program

qAdmesh의 출력은 인터페이스의 메시지 섹션에 표시됩니다. 출력에는 감지된 오류와 출력 옵션이 선택된 경우 이러한 문제점을 해결하기 위해 수행할 조치가 표시됩니다.

사용자 정의 검사 옵션은 파일을 고정할 때 프로그램이 어떤 작업을 수행하는지에 대한 자세한 제어를 제공할 수 있습니다. 또한 변형 및 공차 탭에는 .STL 파일의 회전, 미러링, 크기 조정, 변환 및 병합 기능을 제공하는 옵션이 있습니다.

qAdmesh는 무료 유틸리티입니다만 FSI에서 지원하지 않습니다. qAdmesh가 문제를 해결하는 능력은 심각도에 따라 다릅니다. 문제의 수가 증가함에 따라 qAdmesh 가 문제를 해결할 수 있는 가능성이 줄어 듭니다. 문제를 해결할 수 없는 경우 CAD 패키지를 사용하여  .STL 파일을 재생성 하는 것이 좋습니다.

MiniMagics 

MiniMagics 는 무료 STL파일 시각화 및 복구 유틸리티입니다. 설치는 FLOW-3D 홈 디렉토리 의 Utilites 폴더에서 찾을 수 있으며 파일 분석 및 복구를 위한 유용한 도구로 qAdmesh에서 수행된 수정 사항을 시각화하거나 qAdmesh의 대안으로 사용할 수 있습니다.

$F3D_HOME/UtilitiesSTL

  • Topo2STL

FLOW-3D가 지원하는 유일한 CAD 파일 형식은 .STL이지만 형식을 포함하여 다른 형식의 지형 데이터를 갖는 것은 드문 일이 아닙니다. Topo2STL의 유틸리티로 변환할 수 있습니다. Topo2STL 은 Windows 시스템에서만 사용 가능하며 유틸리티 드롭 다운 메뉴에서 액세스 할 수 있습니다.

명령

  1. 지형 파일은 다음 형식의 ASCII 파일입니다. 각 선은 점을 나타내며 동일한 단위 시스템에서 3 개의 좌표 (일반적으로 피트 또는 미터)를 포함합니다. 좌표는 공백으로 구분됩니다. 선의 좌표 순서는 XYZ 여야 합니다. 여기서 Z는 표고입니다. 두 좌표는 동일한 XY 점을 공유할 수 없습니다. 포인트의 순서 (파일의 줄)는 중요하지 않습니다. 좌표를 포함하지 않는 머리글 줄이나 꼬리 줄이 없어야 합니다.
  2. Topo2stl.exe유틸리티가 추출된 위치에 있는 파일을 실행하여 Topo2STL에 액세스 할 수 있습니다.
  3. 유틸리티를 시작하면 변환할 파일을 선택하라는 topo 파일 찾아보기 창이 나타납니다. 파일 찾아보기 창을 이용하여 파일을 선택합니다.
  4. topo파일이 선택되면, Topo2STL의 창이 나타나고, X, Y의 범위와 Z 계산할 topo데이터 익스텐트가 계산되면 Topo 데이터 익스텐트 및 데이터의 총 포인트 수에 대한 정보가 Information: Topo data extents 아래에 표시됩니다.
Topo2STL
Topo2STL
Topo2STL
Topo2STL
  1. 변환에 필요한 사용자 입력은 공간 분해능 및 STL 최소 Z 좌표입니다. 기본적으로 공간 해상도는 0.002 * min (X 범위, Y 범위)이고 STL 최소 Z 좌표는 ZMIN-(ZMAX-ZMIN)입니다. 여기서 ZMIN 및 ZMAX는 Topo 데이터의 범위입니다.
    • 공간 해상도는 STL 파일을 생성하는 동안 Topo 데이터가 얼마나 정밀하게 분석되는지 제어합니다.
    • STL 최소 Z 좌표는 Topo 데이터의 ZMAX보다 작은 값이어야 합니다. 이것은 STL파일의 최소 ​​Z 두께를 효과적으로 설정합니다.
  2. Browse 버튼은 파일 출력 위치를 설정하는 데 사용할 수 있습니다.
  3. 변환을 클릭하면 변환 프로세스가 시작됩니다. 이 시점에서 변환 취소를 사용하여 변환이 완료되거나 종료될 때까지 Topo2STL 창을 닫을 수 없습니다.
Topo2STL
Topo2STL
  1. 변환이 완료 (또는 종료)되면 변환 단추가 변환 추가로 변경되어 사용자가 변환할 다른 Topo 파일을 선택할 수 있습니다.
Topo2STL
  1. FSAI를 사용한 유한 요소 메쉬 파일 형식 변환

FSAI의 도구에서 유한 요소 메시를 변환하는 유틸리티입니다 Abaqus6.2 이후 형식과 NASTRAN 벌크 형식에 사용되는 형식을 변환하는 FSAI는 유틸리티 드롭 다운 메뉴에서 액세스 할 수 있습니다. FSAI를 사용하려면 다음을 수행하십시오. EXODUS II

  • 적절한 모드에서 유틸리티를 엽니다 (초기 메쉬의 Abaqus 형식인지 NASTRAN 형식인지 여부에 따라 다름 )
  • 파일에서 생성 필드에서 입력 유한 요소 메쉬를 찾습니다.
  • 생성된 파일 위치 필드에서 원하는 출력 위치를 찾으십시오.
  • 생성된 파일 이름 필드에서 원하는 출력 파일 이름을 설정하십시오.
  • 생성을 누릅니다.

 노트

이 FSAI 프로그램을 사용하려면 FLOW-3D 와 별개의 라이센스가 필요합니다. 자세한 내용은 FLOW-3D 영업 담당자에게 문의하십시오.

  1. 계산기

유틸리티 드롭 다운 메뉴에 여러 계산기가 추가되어 알려진 매개 변수 (예: 유체 속성 등)를 기반으로 입력 수량을 추정할 수 있습니다. 사용 가능한 계산기는 다음을 계산합니다.

  • 냉각 채널의 열전달 계수
  • 재료 특성 및 시뮬레이션 시간에 따른 열 침투 깊이
  • 샷 슬리브의 유체 높이
  • 고압 다이캐스팅을 위한 피스톤 속도
  • 밸브 압력 계수
  1. MPDB (Material Properties Database) 확장

MPDB (Material Properties Database)는 FLOW-3D 와 별도로 Flow Science, Inc 에서 구입할 수 있는 타사 데이터베이스입니다. 여기에는 문헌의 다양한 온도 의존성 고체 재료 특성이 포함되어 있습니다. FLOW-3D 용 MPDB는 사용자가 FLOW-3D의 기본 데이터베이스와 호환되는 파일 형식을 내보낼 수 있도록 하여 데이터를 FLOW-3D 로 편리하게 가져올 수 있는 MPDB 독점 버전입니다. MPDB의 재료 특성은 대부분 고체상입니다. 따라서 FLOW-3D 모든 모델 고체 특성을 요구하는 데이터, 특히 유체 구조 상호 작용, 응고 및 열 응력 진화 모델을 활용할 수 있습니다.

MPDB는 다양한 형식으로 데이터를 내보낼 수 있는 독립형 데이터베이스로 사용될 수 있습니다. MPDB에 대한 일반적인 지침은 JAHM Software, Inc.를 방문하십시오. 여기에서는 FLOW-3D 와 함께 MPDB를 사용하는 방법에 대한 지침을 제공합니다. FLOW-3D 와 제대로 통합하려면 MPDB 용 실행 파일이 Windows와 Linux에 있어야 합니다. 실행 파일은 FLOW-3D GUI에 의해 감지되며 재료 메뉴 아래 MPDB에서 재료 가져오기 메뉴 항목 이 활성화됩니다. 이러한 조건 중 하나라도 충족되지 않으면 FLOW-3D GUI를 통해 액세스 할 수 없습니다. MPDB%F3D_HOME%\Utilities$F3D_HOME/UtilitiesMPDB_for_FLOW-3D

FLOW-3D MPDB
FLOW-3D MPDB

material를 클릭 MPDB에서 가져오기 및 사용자 인터페이스 MPDB는 별도의 창에서 열립니다. 재료는 주요 요소로 분류되었습니다. Materials 탭, 테이블에서 요소를 마우스 오른쪽 버튼으로 클릭하여, 사용자는 해당 요소를 포함하는 물질의 목록을 볼 수 있습니다.

(Material Properties Database)
(Material Properties Database)

예를 들어 다음 그림은 철 (Fe)이 포함된 데이터베이스의 재료 목록을 보여줍니다.

FLOW-3D MPDB(Fe)
FLOW-3D MPDB(Fe)

사용자는 다른 합금, 세라믹, 유리 또는 기타 분류되지 않은 재료를 분류하는 다른 탭으로 전환할 수도 있습니다. 다음 그림은 Al & Cu 합금 목록을 보여줍니다.

FLOW-3D MPDB(Al & Cu)
FLOW-3D MPDB(Al & Cu)
FLOW-3D MPDB(Fe,Ni - 1006 (UNS G10060))
FLOW-3D MPDB(Fe,Ni – 1006 (UNS G10060))

재료가 식별되면 재료를 두 번 클릭하면 해당 재료에 사용할 수 있는 속성 목록이 있는 별도의 창이 나타납니다. 예를 들어 Fe 및 Ni 합금에서 1006 (UNS G10060)을 엽니다. 이러한 속성이 모두 FLOW-3D에 사용되는 것은 아닙니다.

FLOW-3D MPDB(1006(UNS G10060))
FLOW-3D MPDB(1006(UNS G10060))

각 속성은 이 창의 오른쪽에서 선택할 수 있는 다른 형식으로 파일에 표시, 플로팅 또는 저장할 수 있습니다. 그러나 이러한 속성 중 일부가 FLOW-3D 로 인식되는 것은 아닙니다. 

FLOW-3D 와 호환되는 파일 형식을 생성하려면 재료 창을 닫고 FLOW-3D/SolidWorks/ANSYS 메뉴에서 시작하십시오. 재료의 특성으로 FLOW-3D로 가져올 수 있는 세 가지 파일 형식이 있습니다.  유체 데이터베이스 형식(.f3d_dbf 확장), 고체 데이터베이스 형식 (.f3d_dbs 확장), 일반 쉼표로 구분된 값(CSV형식)으로 부터 시뮬레이션에 적합한 FLOW-3D 호환 형식을 선택하십시오. MPDB의 재료는 대부분 고체이지만 사용자가 응고된 유체의 특성을 가져오려면 FLOW-3D에서 응고된 유체 특성이 유체 특성의 일부이므로 Fluids 데이터베이스 형식을 선택해야 합니다. 솔리드 및 유체 데이터베이스 파일 형식과 파일은 현재 사용자의 문서 폴더와 Windows 및 Linux에 저장됩니다.

CSV<My Documents>\FLOW-3D\gui\MaterialsDatabase/home/<user>/FLOW-3D/gui/MaterialsDatabase

이러한 위치는 FLOW-3D의 데이터베이스가 사용자 정의 재료를 찾는 곳입니다. MPDB에서 이러한 위치로 내보낸 모든 자료는 FLOW-3D의 기본 데이터베이스에 의해 선택됩니다.

1006 (UNS G10060) 철 합금을 선택하십시오.

FLOW-3D MPDB(UNS G10060)
FLOW-3D MPDB(UNS G10060)

이전에 사용 가능했던 일부 특성은 FLOW-3D 와 관련이 없기 때문에 사용 불가능 합니다. 각 속성이 처리되자 마자 플롯 되거나 해당 데이터가 표시되면 참조 및 메모 섹션이 활성화됩니다. 참조 탭 속성에서 찍은 위치를 나타내는 참고 섹션은 일반적으로 데이터의 구성과 정확성에 관한 사항이 포함되어 있습니다. 

온도에 따른 특성의 동작을 이해하는 데 도움이 되도록 각 특성을 플롯 할 수 있습니다. 또한 데이터의 유효성에 대한 경고가 있을 수 있습니다. 

예를 들어 열전도도를 먼저 플로팅하면 저온 경고가 표시됩니다. 온도의 함수로 플롯을 표시하기 전에 .f3d_dbs파일을 쓰려면 데이터베이스에 추가 버튼을 클릭하고 다음 창에서 파일에 쓸 속성을 ​​선택하십시오. 사용 가능한 단계에 대한 속성을 선택할 수 있습니다. 속성이 선택되면 데이터 쓰기 및 닫기를 클릭하십시오. 

재료 창을 닫습니다. FLOW-3D/SolidWorks/ANSYS 메뉴에서 데이터베이스를 닫습니다.

FLOW-3D MPDB(Low temperature warning)
FLOW-3D MPDB(Low temperature warning)
FLOW-3D MPDB(Temperature Plot)
FLOW-3D MPDB(Temperature Plot)

.f3d_dbs파일을 쓰려면 데이터베이스에 추가 버튼을 클릭하고 다음 창에서 파일에 쓸 속성을 ​​선택하십시오. 사용 가능한 단계에 대한 속성을 선택할 수 있습니다. 속성이 선택되면 데이터 쓰기 및 닫기를 클릭하십시오. 재료 창을 닫습니다. FLOW-3D/SolidWorks/ANSYS 메뉴에서 데이터베이스를 닫습니다.

경우에 따라 재료에 사용자에게 필요한 속성이 없습니다. 데이터베이스에 사용 가능한 속성을 추가한 후 이러한 상황에서 누락된 속성은 유사한 속성을 가진 합금 (사용자의 위험 부담)에서 얻을 수 있습니다. 데이터베이스가 열려있는 동안 FLOW-3D에서 사용될 하나의 재료에 대해 속성을 혼합하고 일치시킬 수 있습니다.

FLOW-3D MPDB(Select properties to write to file)
FLOW-3D MPDB(Select properties to write to file)

데이터베이스를 닫은 후 파일 이름을 묻는 메시지가 사용자에게 표시됩니다. 기본값은 MPDB 가 재료에 지정하는 것입니다. FLOW-3D 가 재료를 사용자 정의 재료로 인식하도록 파일의 위치와 확장자가 미리 설정되어 있습니다.

FLOW-3D MPDB(File locate position)
FLOW-3D MPDB(File locate position)

CSV파일을 선택한 경우에도 동일한 프로세스가 적용됩니다. 데이터가 파일에 기록되면 각 테이블 형식 속성 창의 값 가져오기 버튼에서 데이터를 검색할 수 있습니다.

첫 번째 열은 항상 온도입니다.

FLOW-3D MPDB(csv file)
FLOW-3D MPDB(csv file)
  1. grfedit를 사용하여 flsgrf 파일 편집

명령 줄 유틸리티이므로 runscript와 같은 적절한 환경에서 실행해야 합니다 ( Runscripts 사용 참조 ).


Runscripts 사용

실행 스크립트는 작업 문제 디렉토리에서 실행되도록 설계되었습니다. 스크립트는 $F3D_HOME/local디렉토리에 있습니다. 스크립트를 사용하려면 다음 환경 변수를 설정해야합니다.

  • F3D_HOMEFLOW-3D 설치 디렉터리 의 경로를 지정합니다 .
  • F3DTKNUX_LICENSE_FILEFLOW-3D 라이선스 서버 의 위치를 ​​지정 합니다.
  • PATHPATH포함하도록 환경 변수를 수정해야합니다. $F3D_HOME/local그렇지 않으면 실행 스크립트를 찾을 수 없습니다.
  • F3D_VERSION: 사용할 솔버 버전을 지정합니다. 유효한 옵션은 double배정 밀도 버전 및 prehyd사용자 지정 배정 밀도 솔버입니다.

명령 줄에서 실행하려면 :

  1. 명령 프롬프트 또는 터미널을 엽니 다.
  2. 필요한 환경 변수를 설정하십시오.
    • Windows : FLOW-3D 를 시작하는 데 사용되는 배치 파일에서 환경을 복사하여 수행 할 수 있습니다 . 배치 파일의 내용은 FLOW-3D 아이콘 을 마우스 오른쪽 버튼으로 클릭 하고 편집을 선택 하여 액세스 할 수 있습니다 .
    • Linux : 설치 디렉토리 에서 파일을 flow3dvars.sh가져옵니다 local.
  3. 솔버가 실행중인 디렉토리로 변경하십시오.
  4. 원하는 runscript 명령을 입력하십시오. runhyd <ext2>

  • grfedit를 연 후 사용자에게 소스 파일 (flsgrf.*데이터가 복사될 파일)의 경로를 묻는 메시지가 표시됩니다. 파일의 전체 경로 (예 c:\users\username\FLOW-3D\simulation\flsgrf.simulation:)를 입력하고 <enter>를 누르십시오.
  • 이제, 파일 입력 확장의 목표 예를 들어, (데이터를 기록할 위치로 파일) 파일을 new_output. 데이터가 파일에 기록됩니다 c:\users\username\FLOW-3D\simulation\flsgrf.new_output. 대상 파일이 존재하면 파일을 덮어쓰거나 대상 파일에 데이터를 추가하라는 메시지가 표시됩니다. 대상 파일의 시간보다 늦게 시뮬레이션 시간을 가진 소스 파일 편집 만 추가됩니다.
  • 이 시점에서 프로그램은 어떤 히스토리 데이터 편집, 데이터 편집 재시작 및 대상 파일에 쓰기 위해 선택된 데이터 편집을 묻습니다. 프롬프트에 따라 작성할 데이터 편집을 선택하십시오.
  • 대상 파일을 작성한 후 프로그램이 닫히고 다른 flsgrf.*파일처럼 사용할 수 있습니다.

 노트

  • grfedit는 FLOW-3D v11.1 이상에서 작성된 결과 파일에서만 작동합니다.
  • 소스 flsgrf.*파일은 grfedit에 의해 수정되지 않습니다
  • FLOW-3D/MP의 출력 파일로 작업할 때는 flsgrf1의 위로 flsgrf 교체 하십시오 .
  • 소스 및 대상 파일 모두에 허용되는 유일한 이름은 flsgrf및 flsgrf1입니다.

FLOW-3D 및TruVOF는 미국 및 기타 국가에서 등록 상표입니다.

FLOW-3D 용어 사전 테이블

FLOW-3D Glossary | FLOW-3D 용어 사전

FLOW-3D 용어 사전 / 용어 설명

FLOW-3D 용어 사전 테이블
FLOW-3D 용어 사전 테이블

FLOW-3D 용어 사전 / 용어 설명

Drift Flux

드리프트 모델은 밀도가 서로 다른 두 혼합 유체 구성 요소의 상대적 흐름을 설명합니다. 구성 요소는 상이 다를 수도 있고, 상이 같지만(불가침) 유체가 다를 수도 있습니다. 분산된 위상 입자 크기가 클 경우 드리프트 모델의 적용성에 대한 제한이 존재할 수 있습니다. 이러한 제한은 일반적으로 메쉬 셀 크기의 10% 미만으로 분산된 위상 입자 크기를 유지함으로써 방지할 수 있습니다.

배플

얇은 형상 조각을 나타내는데 사용되는 2 차원 개체입니다. 이들은 전처리기에 의해 셀면으로 이동되고 유체의 흐름을 부분적으로 또는 완전히 차단하는 역할을 합니다. 배플은 지정된 열 전달 계수를 가질 수 있으며 배플을 통과하는 양(플럭스 표면)을 측정하는 데 사용할 수 있습니다.

Two-dimensional objects that are used to represent thin pieces of geometry. They are moved by the preprocessor to cell faces and act to partially, or completely block the flow of fluid. Baffles can have heat transfer coefficients specified and can be used to measure quantities that pass through them (a flux surface).

경계 조건

도메인의 범위에서 솔루션을 정의합니다. 경계 위치에서 흐름의 실제 상태를 나타내는 경계 조건을 선택하는 것이 중요합니다.

Defines the solution at the extents of the domain. It is important to choose boundary conditions that represent the true condition of the flow at the boundary location.

CFD

CFD (Computational Fluid Dynamics)는 수치 솔루션을 통해 컴퓨터의 유체 흐름을 시뮬레이션 하는 유체 역학의 한 분야입니다.

Computational Fluid Dynamics (CFD), the branch of fluid mechanics dedicated to simulating the flow of fluid on a computer via numerical solutions.

Complements

Complements를 정의합니다. 예를 들어, 솔리드 구의 complements는 솔리드 재료로 둘러싸인 구형 구멍입니다.

The inverse of a shape defines the complement. For example, the complement of a solid sphere is a spherical hole surrounded by solid material.

Client

클라이언트 컴퓨터는 자신이 FLOW-3D를 실행하고 있지만, FLOW-3D 소프트웨어 라이선스는 다른 컴퓨터 (서버 컴퓨터)에서 획득하는 컴퓨터를 의미합니다.

A client machine is a computer that runs FLOW-3D  but acquires the software license from a different machine (the server machine)

Components

Components는 공간의 개체를 정의하며 하위 구성 요소로 구성됩니다. 구성 요소는 열 전도율, 비열 및 표면 거칠기와 같은 재료 특성을 가질 수 있습니다.

Components define objects in space and are comprised of subcomponents. A component can have material properties such as thermal conductivity, specific heat and surface roughness.

Custom result

시뮬레이션 중 또는 완료 후 사용자가 생성한 데이터를 그래픽으로 표시합니다. 생성하려면 사용자가 flsgrf결과 파일을 연 다음 플로팅 매개 변수(예 : 플로팅 할 도메인 부분, 플로팅 할 수량 등)를 선택해야 합니다.

Graphical displays of data generated by the user during the simulation or after it has completed. To generate, the user must open an flsgrf results file and then select the plotting parameter (e.g., portion of domain to plot, quantity to plot, etc.).

Domain

지배 방정식을 풀 영역입니다. 이것은 메쉬의 범위에 의해 정의됩니다.

The region in which the governing equations are to be solved. This is defined by the extents of the mesh.

Diagnostics

전 처리기 및 솔버의 진행 상황과 오류 및 경고에 대한 정보가 포함된 파일 세트입니다.

A suite of files that contain information on the progress of the preprocessor and solver as well as errors and warnings.

EPSI

압력/연속 반복이 어느 지점에서 수렴되는지를 결정하는데 사용된 수렴 기준입니다. 기본 숫자 설정을 사용하면 이 값은 FLOW-3D에 의해 자동으로 계산되며 시간 단계가 증가함에 따라 작아집니다.

The convergence criterion that was used to determine at what point the pressure/continuity iterations have converged. With the default numerical settings, this value is automatically computed by FLOW-3D  and becomes smaller as the time step increases.

Existing result

prpplt.* 또는 flsplt.* 파일은 전처리 종료 솔버 실행 종료시 또는 자동으로 생성되는 플롯 파일입니다.

A plot file that is automatically created, either at the end of preprocessing or the end of the solver run- prpplt.* or flsplt.*.

F3D_HOME

FLOW-3D 프로그램 파일이 있는 디렉토리를 정의하는 환경 변수.

Environment variable that defines the directory where the FLOW-3D  program files are located.

Floating license

FLOW-3D는 서버 시스템에 라이센스를 액세스하는 각 클라이언트 컴퓨터와 컴퓨터 네트워크에서 실행합니다. 허용하는 라이센스 최대 동시 시뮬레이션 수는 구매한 솔버 토큰 수에 의해 제한됩니다.

A license that allows FLOW-3D  to be run on a network of computers with each client machine accessing the license on a server machine. The maximum number of concurrent simulations is limited by the number of solver