Figure 4.24 - Model with virtual valves in the extremities of the geometries to simulate the permeability of the mold promoting a more uniformed filling

Optimization of filling systems for low pressure by Flow-3D

Dissertação de Mestrado
Ciclo de Estudos Integrados Conducentes ao
Grau de Mestre em Engenharia Mecânica
Trabalho efectuado sob a orientação do
Doutor Hélder de Jesus Fernades Puga
Professor Doutor José Joaquim Carneiro Barbosa

ABSTRACT

논문의 일부로 튜터 선택 가능성과 해결해야 할 주제가 설정되는 매개변수를 염두에 두고 개발 주제 ‘Flow- 3D ®에 의한 저압 충전 시스템 최적화’가 선택되었습니다. 이를 위해서는 달성해야 할 목표와 이를 달성하기 위한 방법을 정의하는 것이 필요했습니다.

충전 시스템을 시뮬레이션하고 검증할 수 있는 광범위한 소프트웨어에도 불구하고 Flow-3D®는 시장에서 최고의 도구 중 하나로 표시되어 전체 충전 프로세스 및 행동 표현과 관련하여 탁월한 정확도로 시뮬레이션하는 능력을 입증했습니다.

이를 위해 관련 프로세스를 더 잘 이해하고 충진 시스템 시뮬레이션을 위한 탐색적 기반 역할을 하기 위해 이 도구를 탐색하는 것이 중요합니다. 지연 및 재료 낭비에 반영되는 실제적인 측면에서 충전 장치의 치수를 완벽하게 만드는 비용 및 시간 낭비. 이러한 방식으로 저압 주조 공정에서 충진 시스템을 설계하고 물리적 모델을 탐색하여 특성화하는 방법론을 검증하기 위한 것입니다.

이를 위해 다음 주요 단계를 고려하십시오.

시뮬레이션 소프트웨어 Flow 3D® 탐색;
충전 시스템 모델링;
모델의 매개변수를 탐색하여 모델링된 시스템의 시뮬레이션, 검증 및 최적화.

따라서 연구 중인 압력 곡선과 주조 분석에서 가장 관련성이 높은 정보의 최종 마이닝을 검증하기 위한 것입니다.

사용된 압력 곡선은 수집된 문헌과 이전에 수행된 실제 작업을 통해 얻었습니다. 결과를 통해 3단계 압력 곡선이 층류 충진 체계의 의도된 목적과 관련 속도가 0.5 𝑚/𝑠를 초과하지 않는다는 결론을 내릴 수 있었습니다.

충전 수준이 2인 압력 곡선은 0.5 𝑚/𝑠 이상의 속도로 영역을 채우는 더 난류 시스템을 갖습니다. 열전달 매개변수는 이전에 얻은 값이 주물에 대한 소산 거동을 확증하지 않았기 때문에 연구되었습니다.

이러한 방식으로 주조 공정에 더 부합하는 새로운 가치를 얻었습니다. 달성된 결과는 유사한 것으로 나타난 NovaFlow & Solid®에 의해 생성된 결과와 비교되어 시뮬레이션에서 설정된 매개변수를 검증했습니다. Flow 3D®는 주조 부품 시뮬레이션을 위한 강력한 도구로 입증되었습니다.

As part of the dissertation and bearing in mind the parameters in which the possibility of a choice of tutor and the subject to be addressed is established, the subject for development ’Optimization of filling systems for low pressure by Flow 3D ®’ was chosen. For this it was necessary to define the objectives to achieve and the methods to attain them. Despite the wide range of software able to simulate and validate filling systems, Flow 3D® has been shown as one of the best tools in the market, demonstrating its ability to simulate with distinctive accuracy with respect to the entire process of filling and the behavioral representation of the fluid obtained. To this end, it is important to explore this tool for a better understanding of the processes involved and to serve as an exploratory basis for the simulation of filling systems, simulation being one of the great strengths of the current industry due to the need to reduce costs and time waste, in practical terms, that lead to the perfecting of the dimensioning of filling devices, which are reflected in delays and wasted material. In this way it is intended to validate the methodology to design a filling system in lowpressure casting process, exploring their physical models and thus allowing for its characterization. For this, consider the following main phases: The exploration of the simulation software Flow 3D®; modeling of filling systems; simulation, validation and optimization of systems modeled by exploring the parameters of the models. Therefore, it is intended to validate the pressure curves under study and the eventual mining of the most relevant information in a casting analysis. The pressure curves that were used were obtained through the gathered literature and the practical work previously performed. Through the results it was possible to conclude that the pressure curve with 3 levels meets the intended purpose of a laminar filling regime and associated speeds never exceeding 0.5 𝑚/𝑠. The pressure curve with 2 filling levels has a more turbulent system, having filling areas with velocities above 0.5 𝑚/𝑠. The heat transfer parameter was studied due to the values previously obtained didn’t corroborate the behavior of dissipation regarding to the casting. In this way, new values, more in tune with the casting process, were obtained. The achieved results were compared with those generated by NovaFlow & Solid®, which were shown to be similar, validating the parameters established in the simulations. Flow 3D® was proven a powerful tool for the simulation of casting parts.

키워드

저압, Flow 3D®, 시뮬레이션, 파운드리, 압력-시간 관계,Low Pressure, Flow 3D®, Simulation, Foundry, Pressure-time relation

Figure 4.24 - Model with virtual valves in the extremities of the geometries to simulate the permeability of the mold promoting a more uniformed filling
Figure 4.24 – Model with virtual valves in the extremities of the geometries to simulate the permeability of the mold promoting a more uniformed filling
Figure 4.39 - Values of temperature contours using full energy heat transfer parameter for simula
Figure 4.39 – Values of temperature contours using full energy heat transfer parameter for simula
Figure 4.40 – Comparison between software simulations (a) Flow 3D® simulation,
(b) NovaFlow & Solid® simulation
Figure 4.40 – Comparison between software simulations (a) Flow 3D® simulation, (b) NovaFlow & Solid® simulation

BIBLIOGRAPHY

[1] E. Stanley and D. B. Sc, “Fluid Flow Aspects of Solidification Modelling : Simulation
of Low Pressure Die Casting .”
[2] Y. Sahin, “Computer aided foundry die-design,” Metallography, vol. 24, no. 8, pp.
671–679, 2003.
[3] F. Bonollo, J. Urban, B. Bonatto, and M. Botter, “Gravity and low pressure die casting
of aluminium alloys : a technical and economical benchmark,” La Metall. Ital., vol. 97,
no. 6, pp. 23–32, 2005.
[4] P. a and R. R, “Study of the effect of process parameters on the production of a nonsimmetric low pressure die casting part,” La Metall. Ital., pp. 57–63, 2009.
[5] “Fundição em baixa pressão | Aluinfo.” [Online]. Available:
http://www.aluinfo.com.br/novo/materiais/fundicao-em-baixa-pressao. [Accessed: 18-
Sep-2015].
[6] “Low Pressure Sand Casting by Wolverine Bronze.” [Online]. Available:
http://www.wolverinebronze.com/low-pressure-sand-casting.php. [Accessed: 18-Sep2015].
[7] A. Reikher, “Numerical Analysis of Die-Casting Process in Thin Cavities Using
Lubrication Approximation,” no. December, 2012.
[8] P. Fu, A. a. Luo, H. Jiang, L. Peng, Y. Yu, C. Zhai, and A. K. Sachdev, “Low-pressure
die casting of magnesium alloy AM50: Response to process parameters,” J. Mater.
Process. Technol., vol. 205, no. 1–3, pp. 224–234, 2008.
[9] X. Li, Q. Hao, W. Jie, and Y. Zhou, “Development of pressure control system in
counter gravity casting for large thin-walled A357 aluminum alloy components,”
Trans. Nonferrous Met. Soc. China, vol. 18, no. 4, pp. 847–851, 2008.
[10] J. a. Hines, “Determination of interfacial heat-transfer boundary conditions in an
aluminum low-pressure permanent mold test casting,” Metall. Mater. Trans. B, vol. 35,
no. 2, pp. 299–311, 2004.
[11] A. Lima, A. Freitas, and P. Magalhães, “Processos de vazamento em moldações
permanentes,” pp. 40–49, 2003.
[12] Y. B. Choi, K. Matsugi, G. Sasaki, K. Arita, and O. Yanagisawa, “Analysis of
Manufacturing Processes for Metal Fiber Reinforced Aluminum Alloy Composite
Fabricated by Low-Pressure Casting,” Mater. Trans., vol. 47, no. 4, pp. 1227–1231,
68
2006.
[13] G. Mi, X. Liu, K. Wang, and H. Fu, “Numerical simulation of low pressure die-casting
aluminum wheel,” China Foundry, vol. 6, no. 1, pp. 48–52, 2009.
[14] J. Kuo, F. Hsu, and W. Hwang, “ADVANCED Development of an interactive
simulation system for the determination of the pressure ± time relationship during the
® lling in a low pressure casting process,” vol. 2, pp. 131–145, 2001.
[15] S.-G. Liu, F.-Y. Cao, X.-Y. Zhao, Y.-D. Jia, Z.-L. Ning, and J.-F. Sun, “Characteristics
of mold filling and entrainment of oxide film in low pressure casting of A356 alloy,”
Mater. Sci. Eng. A, vol. 626, pp. 159–164, 2015.
[16] “Casting Training Class – Lecture 10 – Solidification and Shrinkage-Casting.” FLOW3D®.
[17] “UAB Casting Engineering Laboratory.” [Online]. Available:
file:///C:/Users/Jos%C3%A9 Belo/Desktop/Artigo_Software/UAB Casting
Engineering Laboratory.htm. [Accessed: 09-Nov-2015].
[18] A. Louvo, “Casting Simulation as a Tool in Concurrent Engineering,” pp. 1–12, 1997.
[19] T. R. Vijayaram and P. Piccardo, “Computers in Foundries,” vol. 30, 2012.
[20] M. Sadaiah, D. R. Yadav, P. V. Mohanram, and P. Radhakrishnan, “A generative
computer-aided process planning system for prismatic components,” Int. J. Adv.
Manuf. Technol., vol. 20, no. 10, pp. 709–719, 2002.
[21] Ministry_of_Planning, “Digital Data,” vol. 67, pp. 1–6, 2004.
[22] S. Shamasundar, D. Ramachandran, and N. S. Shrinivasan, “COMPUTER
SIMULATION AND ANALYSIS OF INVESTMENTCASTING PROCESS.”
[23] J. M. Siqueira and G. Motors, “Simulation applied to Aluminum High Pressure Die
Casting,” pp. 1–5, 1998.
[24] C. Fluid, COMPUTATIONAL FLUID DYNAMICS. Abdulnaser Sayma & Ventus
Publishing ApS, 2009.
[25] C. a. Felippa, “1 – Overview,” Adv. Finite Elem. Methods, pp. 1–9.
[26] a. Meena and M. El Mansori, “Correlative thermal methodology for castability
simulation of ductile iron in ADI production,” J. Mater. Process. Technol., vol. 212,
no. 11, pp. 2484–2495, 2012.
[27] T. R. Vijayaram, S. Sulaiman, a. M. S. Hamouda, and M. H. M. Ahmad, “Numerical
simulation of casting solidification in permanent metallic molds,” J. Mater. Process.
69
Technol., vol. 178, pp. 29–33, 2006.
[28] “General CFD FAQ — CFD-Wiki, the free CFD reference.” [Online]. Available:
http://www.cfd-online.com/Wiki/General_CFD_FAQ. [Accessed: 10-Nov-2015].
[29] “FEM | FEA | CFD.” [Online]. Available: http://fem4analyze.blogspot.pt/. [Accessed:
09-Nov-2015].
[30] “Fundição; revista da Associação portuguesa de fundição,” Fundição, vol. N
o
227.
[31] “Casting Training Class – Lecture 1 – Introduction_to_FLOW-3D – Casting.” FLOW3D®.
[32] F. Science, “FLOW-3D Cast Documentation,” no. 3.5, p. 80, 2012.
[33] “Casting Training Class – Lecture 4 – Geometry Building – General.” FLOW-3D®.
[34] F. Science, “FLOW-3D v11.0.3 User Manual,” pp. 1–132, 2015.
[35] “Casting Training Class – Lecture 5 Meshing Concept – General.” FLOW-3D®.
[36] “Casting Training Class – Lecture 6 – Boundary_Conditions – Casting.” FLOW-3D®.
[37] “Casting Training Class – Lecture 9 – Physical Models-castings.” FLOW-3D®.
[38] P. A. D. Jácome, M. C. Landim, A. Garcia, A. F. Furtado, and I. L. Ferreira, “The
application of computational thermodynamics and a numerical model for the
determination of surface tension and Gibbs–Thomson coefficient of aluminum based
alloys,” Thermochim. Acta, vol. 523, no. 1–2, pp. 142–149, 2011.
[39] J. P. Anson, R. A. L. Drew, and J. E. Gruzleski, “The surface tension of molten
aluminum and Al-Si-Mg alloy under vacuum and hydrogen atmospheres,” Metall.
Mater. Trans. B Process Metall. Mater. Process. Sci., vol. 30, no. 6, pp. XVI–1032,
1999.

Figure 1: Mold drawings

3D Flow and Temperature Analysis of Filling a Plutonium Mold

플루토늄 주형 충전의 3D 유동 및 온도 분석

Authors: Orenstein, Nicholas P. [1]

Publication Date:2013-07-24
Research Org.: Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.: DOE/LANL
OSTI Identifier: 1088904
Report Number(s): LA-UR-13-25537
DOE Contract Number: AC52-06NA25396
Resource Type: Technical Report
Country of Publication: United States
Language: English
Subject: Engineering(42); Materials Science(36); Radiation Chemistry, Radiochemistry, & Nuclear Chemistry(38)

Introduction

The plutonium foundry at Los Alamos National Laboratory casts products for various special nuclear applications. However, plutonium’s radioactivity, material properties, and security constraints complicate the ability to perform experimental analysis of mold behavior. The Manufacturing Engineering and Technologies (MET-2) group previously developed a graphite mold to vacuum cast small plutonium disks to be used by the Department of Homeland Security as point sources for radiation sensor testing.

A two-stage pouring basin consisting of a funnel and an angled cavity directs the liquid into a vertical runner. A stack of ten disk castings connect to the runner by horizontal gates. Volumetric flow rates were implemented to limit overflow into the funnel and minimize foundry returns. Models using Flow-3D computational fluid dynamics software are employed here to determine liquid Pu flow paths, optimal pour regimes, temperature changes, and pressure variations.

Setup

Hardcopy drawings provided necessary information to create 3D .stl models for import into Flow-3D (Figs. 1 and 2). The mesh was refined over several iterations to isolate the disk cavities, runner, angled cavity, funnel, and input pour. The final flow and mold-filling simulation utilizes a fine mesh with ~5.5 million total cells. For the temperature study, the mesh contained 1/8 as many cells to reduce computational time and set temperatures to 850 °C for the molten plutonium and 500 °C for the solid graphite mold components (Fig. 3).

Flow-3D solves mass continuity and Navier-Stokes momentum equations over the structured rectangular grid model using finite difference and finite volume numerical algorithms. The solver includes terms in the momentum equation for body and viscous accelerations and uses convective heat transfer.

Simulation settings enabled Flow-3D physics calculations for gravity at 980.665 cm/s 2 in the negative Z direction (top of mold to bottom); viscous, turbulent, incompressible flow using dynamically-computed Renormalized Group Model turbulence calculations and no-slip/partial slip wall shear, and; first order, full energy equation heat transfer.

Mesh boundaries were all set to symmetric boundary conditions except for the Zmin boundary set to outflow and the Zmax boundary set to a volume flow. Vacuum casting conditions and the high reactivity of remaining air molecules with Pu validate the assumption of an initially fluidless void.

Results

The flow follows a unique three-dimensional path. The mold fills upwards with two to three disks receiving fluid in a staggered sequence. Figures 5-9 show how the fluid fills the cavity, and Figure 7 includes the color scale for pressure levels in these four figures. The narrow gate causes a high pressure region which forces the fluid to flow down the cavity centerline.

It proceeds to splash against the far wall and then wrap around the circumference back to the gate (Figs. 5 and 6). Flow in the angled region of the pouring basin cascades over the bottom ledge and attaches to the far wall of the runner, as seen in Figure 7.

This channeling becomes less pronounced as fluid volume levels increase. Finally, two similar but non-uniform depressed regions form about the centerline. These regions fill from their perimeter and bottom until completion (Fig. 8). Such a pattern is counter, for example, to a steady scenario in which a circle of molten Pu encompassing the entire bottom surface rises as a growing cylinder.

Cavity pressure becomes uniform when the cavity is full. Pressure levels build in the rising well section of the runner, where impurities were found to settle in actual casting. Early test simulations optimized the flow as three pours so that the fluid would never overflow to the funnel, the cavities would all fill completely, and small amounts of fluid would remain as foundry returns in the angled cavity.

These rates and durations were translated to the single 2.7s pour at 100 cm 3 per second used here. Figure 9 shows anomalous pressure fluctuations which occurred as the cavities became completely filled. Multiple simulations exhibited a rapid change in pressure from positive to negative and back within the newly-full disk and surrounding, already-full disks.

The time required to completely fill each cavity is plotted in Figure 10. Results show negligible temperature change within the molten Pu during mold filling and, as seen in Figure 11, at fill completion.

Figure 1: Mold drawings
Figure 1: Mold drawings
Figure 2: Mold Assembly
Figure 2: Mold Assembly
Figure 4: Actual mold and cast Pu
Figure 4: Actual mold and cast Pu
Figure 5: Bottom cavity filling
from runner
Figure 5: Bottom cavity filling from runner
Figure 6: Pouring and filling
Figure 6: Pouring and filling
Figure 8: Edge detection of cavity fill geometry. Two similar depressed areas form
about the centerline. Top cavity shown; same pressure scale as other figures
Figure 8: Edge detection of cavity fill geometry. Two similar depressed areas form about the centerline. Top cavity shown; same pressure scale as other figures
Figure 10: Cavity fill times,from first fluid contact with pouring basin, Figure 11:Fluid temperature remains essentially constant
Figure 10: Cavity fill times,from first fluid contact with pouring basin, Figure 11:Fluid temperature remains essentially constant

Conclusions

Non-uniform cavity filling could cause crystal microstructure irregularities during solidification. However, the small temperature changes seen – due to large differences in specific heat between Pu and graphite – over a relatively short time make such problems unlikely in this case.

In the actual casting, cooling required approximately ten minutes. This large difference in time scales further reduces the chance for temperature effects in such a superheated scenario. Pouring basin emptying decreases pressure at the gate which extends fill time of the top two cavities.

The bottom cavity takes longer to fill because fluid must first enter the runner and fill the well. Fill times continue linearly until the top two cavities. The anomalous pressure fluctuations may be due to physical attempts by the system to reach equilibrium, but they are more likely due to numerical errors in the Flow3D solver.

Unsuccessful tests were performed to remove them by halving fluid viscosity. The fine mesh reduced, but did not eliminate, the extent of the fluctuations. Future work is planned to study induction and heat transfer in the full Pu furnace system, including quantifying temporal lag of the cavity void temperature to the mold wall temperature during pre-heat and comparing heat flux levels between furnace components during cool-down.

Thanks to Doug Kautz for the opportunity to work with MET-2 and for assigning an interesting unclassified project. Additional thanks to Mike Bange for CFD guidance, insight of the project’s history, and draft review.

Fig. 6 LH2 isotherms at 1020 s.

액체-수소 탱크를 위한 결합된 열역학-유체-역학 솔루션

Coupled thermodynamic-fluid-dynamic solution for a liquid-hydrogen tank

G. D. Grayson

Published Online:23 May 2012 https://doi.org/10.2514/3.26706

Tools Share

Free first page

Introduction

ROPELLANT 열 성층화 및 외부 교란에 대한 유체 역학적 반응은 발사체와 우주선 모두에서 중요합니다. 과거에는 결합된 솔루션을 제공할 수 있는 충분한 계산 기술이 부족하여 이러한 문제를 개별적으로 해결했습니다.1

이로 인해 모델링 기술의 불확실성을 허용하기 위해 큰 안전 계수를 가진 시스템이 과도하게 설계되었습니다. 고중력 환경과 저중력 환경 모두에서 작동하도록 설계된 미래 시스템은 기술적으로나 재정적으로 실현 가능하도록 과잉 설계 및 안전 요소가 덜 필요합니다.

이러한 유체 시스템은 열역학 및 유체 역학이 모두 중요한 환경에서 모델의 기능을 광범위하게 검증한 후에만 고충실도 수치 모델을 기반으로 할 수 있습니다. 상용 컴퓨터 코드 FLOW-3D2는 유체 역학 및 열 모델링 모두에서 가능성을 보여주었으며,1 따라서 열역학-유체-역학 엔지니어링 문제에서 결합된 질량, 운동량 및 에너지 방정식을 푸는 데 적합함을 시사합니다.

발사체의 복잡한 액체 가스 시스템에 대한 포괄적인 솔루션을 달성하기 위한 첫 번째 단계로 액체 유체 역학과 열역학을 통합하는 제안된 상단 단계 액체-수소(Lit) 탱크의 간단한 모델이 여기에 제시됩니다. FLOW-3D FLOW-3D 프로그램은 Los Alamos Scientific Laboratory에서 시작되었으며 마커 및 셀 방법에서 파생된 것입니다.3 현재 상태로 가져오기 위해 수년에 걸쳐 광범위한 코드 수정이 이루어졌습니다.2

프로그램은 다음과 같습니다. 일반 Navier-Stokes 방정식을 풀기 위해 수치 근사의 중앙 유한 차분 방법을 사용하는 3차원 유체 역학 솔버입니다. 모멘텀 및 에너지 방정식의 섹션은 특정 응용 프로그램에 따라 활성화 또는 비활성화할 수 있습니다.

코드는 1994년 9월 13일 접수를 인용하기 위해 무액체 표면, 복잡한 용기 기하학, 여러 점성 모델, 표면 장력, 다공성 매체를 통한 흐름 및 응고와 함께 압축성 또는 비압축성 유동 가정을 제공합니다. 1995년 1월 15일에 받은 개정; 1995년 2월 17일 출판 승인.

ROPELLANT thermal stratification and fluid-dynamic response to external disturbances are of concern in both launch vehicles and spacecraft. In the past these problems have been addressed separately for want of sufficient computational technology to provide for coupled solutions.1 This has resulted in overdesigned systems with large safety factors to allow for the uncertainty in modeling techniques. Future systems designed to perform in both highand low-gravity environments will require less overdesign and safety factors to be technically and financially feasible. Such fluid systems can be based on high-fidelity numerical models only after extensive validation of the models’ capabilities in environments where both the thermodynamics and the fluid dynamics are important. The commercial computer code FLOW-3D2 has shown promise in both fluid-dynamic and thermal modeling,1 thus suggesting suitability for solving the coupled mass, momentum, and energy equations in thermodynamic-fluid-dynamic engineering problems. As a first step to achieving a comprehensive solution for complex liquidgas systems in a launch vehicle, a simple model of a proposed upper-stage liquid-hydrogen (Lit) tank incorporating the liquid fluid dynamics and thermodynamics is presented here. FLOW-3D The FLOW-3D program originated at the Los Alamos Scientific Laboratory and is a derivative of the marker-and-cell method.3 Extensive code modifications have been made over the years to bring it to its present state.2 The program is a three-dimensional fluiddynamic solver that uses a central finite-difference method of numerical approximation to solve the general Navier-Stokes equations. Sections of the momentum and energy equations can be enabled or disabled depending on the particular application. The code provides compressible or incompressible flow assumptions with liquid free surfaces, complex container geometries, several viscosity models, surface tension, flow though porous media, and solidification, to cite Received Sept. 13, 1994; revision received Jan. 15, 1995; accepted for publication Feb. 17, 1995. Copyright © 1995 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. *Engineer/Scientist, Propulsion Analysis and Hydraulics, Space Transportation Division, MS 13-3, 5301 Bolsa Avenue. Member AIAA. a few of the possibilities. Further information on FLOW-3D’s capabilities and details of the numerical algorithms can be found in Ref. 2

Fig. 1 Axial-acceleration history.
Fig. 1 Axial-acceleration history.
Fig. 2 Heat flux histories.
Fig. 2 Heat flux histories.
Fig. 3 LHi isotherms at 50 s.
Fig. 3 LHi isotherms at 50 s.
Fig. 4 LH2 isotherms at 300 s
Fig. 4 LH2 isotherms at 300 s
Fig. 5 LH2 isotherms at 880 s.
Fig. 5 LH2 isotherms at 880 s.
Fig. 6 LH2 isotherms at 1020 s.
Fig. 6 LH2 isotherms at 1020 s.
Fig. 7 Tank-outlet temperature history.
Fig. 7 Tank-outlet temperature history.
Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling

영국 Dawlish의 방파제에 대한 온대 저기압 피해: 목격자 설명, 해수면 분석 및 수치 모델링

Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling

Natural Hazards (2022)Cite this article

Abstract

2014년 2월 영국 해협(영국)과 특히 Dawlish에 영향을 미친 온대 저기압 폭풍 사슬은 남서부 지역과 영국의 나머지 지역을 연결하는 주요 철도에 심각한 피해를 입혔습니다.

이 사건으로 라인이 두 달 동안 폐쇄되어 5천만 파운드의 피해와 12억 파운드의 경제적 손실이 발생했습니다. 이 연구에서는 폭풍의 파괴력을 해독하기 위해 목격자 계정을 수집하고 해수면 데이터를 분석하며 수치 모델링을 수행합니다.

우리의 분석에 따르면 이벤트의 재난 관리는 성공적이고 효율적이었으며 폭풍 전과 도중에 인명과 재산을 구하기 위해 즉각적인 조치를 취했습니다. 파도 부이 분석에 따르면 주기가 4–8, 8–12 및 20–25초인 복잡한 삼중 봉우리 바다 상태가 존재하는 반면, 조위계 기록에 따르면 최대 0.8m의 상당한 파도와 최대 1.5m의 파도 성분이 나타났습니다.

이벤트에서 가능한 기여 요인으로 결합된 진폭. 최대 286 KN의 상당한 임펄스 파동이 손상의 시작 원인일 가능성이 가장 높았습니다. 수직 벽의 반사는 파동 진폭의 보강 간섭을 일으켜 파고가 증가하고 최대 16.1m3/s/m(벽의 미터 너비당)의 상당한 오버탑핑을 초래했습니다.

이 정보와 우리의 공학적 판단을 통해 우리는 이 사고 동안 다중 위험 계단식 실패의 가장 가능성 있는 순서는 다음과 같다고 결론을 내립니다. 조적 파괴로 이어지는 파도 충격력, 충전물 손실 및 연속적인 조수에 따른 구조물 파괴.

The February 2014 extratropical cyclonic storm chain, which impacted the English Channel (UK) and Dawlish in particular, caused significant damage to the main railway connecting the south-west region to the rest of the UK. The incident caused the line to be closed for two months, £50 million of damage and an estimated £1.2bn of economic loss. In this study, we collate eyewitness accounts, analyse sea level data and conduct numerical modelling in order to decipher the destructive forces of the storm. Our analysis reveals that the disaster management of the event was successful and efficient with immediate actions taken to save lives and property before and during the storm. Wave buoy analysis showed that a complex triple peak sea state with periods at 4–8, 8–12 and 20–25 s was present, while tide gauge records indicated that significant surge of up to 0.8 m and wave components of up to 1.5 m amplitude combined as likely contributing factors in the event. Significant impulsive wave force of up to 286 KN was the most likely initiating cause of the damage. Reflections off the vertical wall caused constructive interference of the wave amplitudes that led to increased wave height and significant overtopping of up to 16.1 m3/s/m (per metre width of wall). With this information and our engineering judgement, we conclude that the most probable sequence of multi-hazard cascading failure during this incident was: wave impact force leading to masonry failure, loss of infill and failure of the structure following successive tides.

Introduction

The progress of climate change and increasing sea levels has started to have wide ranging effects on critical engineering infrastructure (Shakou et al. 2019). The meteorological effects of increased atmospheric instability linked to warming seas mean we may be experiencing more frequent extreme storm events and more frequent series or chains of events, as well as an increase in the force of these events, a phenomenon called storminess (Mölter et al. 2016; Feser et al. 2014). Features of more extreme weather events in extratropical latitudes (30°–60°, north and south of the equator) include increased gusting winds, more frequent storm squalls, increased prolonged precipitation and rapid changes in atmospheric pressure and more frequent and significant storm surges (Dacre and Pinto 2020). A recent example of these events impacting the UK with simultaneous significant damage to coastal infrastructure was the extratropical cyclonic storm chain of winter 2013/2014 (Masselink et al. 2016; Adams and Heidarzadeh 2021). The cluster of storms had a profound effect on both coastal and inland infrastructure, bringing widespread flooding events and large insurance claims (RMS 2014).

The extreme storms of February 2014, which had a catastrophic effect on the seawall of the south Devon stretch of the UK’s south-west mainline, caused a two-month closure of the line and significant disruption to the local and regional economy (Fig. 1b) (Network Rail 2014; Dawson et al. 2016; Adams and Heidarzadeh 2021). Restoration costs were £35 m, and economic effects to the south-west region of England were estimated up to £1.2bn (Peninsula Rail Taskforce 2016). Adams and Heidarzadeh (2021) investigated the disparate cascading failure mechanisms which played a part in the failure of the railway through Dawlish and attempted to put these in the context of the historical records of infrastructure damage on the line. Subsequent severe storms in 2016 in the region have continued to cause damage and disruption to the line in the years since 2014 (Met Office 2016). Following the events of 2014, Network Rail Footnote1 who owns the network has undertaken a resilience study. As a result, it has proposed a £400 m refurbishment of the civil engineering assets that support the railway (Fig. 1) (Network Rail 2014). The new seawall structure (Fig. 1a,c), which is constructed of pre-cast concrete sections, encases the existing Brunel seawall (named after the project lead engineer, Isambard Kingdom Brunel) and has been improved with piled reinforced concrete foundations. It is now over 2 m taller to increase the available crest freeboard and incorporates wave return features to minimise wave overtopping. The project aims to increase both the resilience of the assets to extreme weather events as well as maintain or improve amenity value of the coastline for residents and visitors.

figure 1
Fig. 1

In this work, we return to the Brunel seawall and the damage it sustained during the 2014 storms which affected the assets on the evening of the 4th and daytime of the 5th of February and eventually resulted in a prolonged closure of the line. The motivation for this research is to analyse and model the damage made to the seawall and explain the damage mechanisms in order to improve the resilience of many similar coastal structures in the UK and worldwide. The innovation of this work is the multidisciplinary approach that we take comprising a combination of analysis of eyewitness accounts (social science), sea level and wave data analysis (physical science) as well as numerical modelling and engineering judgement (engineering sciences). We investigate the contemporary wave climate and sea levels by interrogating the real-time tide gauge and wave buoys installed along the south-west coast of the English Channel. We then model a typical masonry seawall (Fig. 2), applying the computational fluid dynamics package FLOW3D-Hydro,Footnote2 to quantify the magnitude of impact forces that the seawall would have experienced leading to its failure. We triangulate this information to determine the probable sequence of failures that led to the disaster in 2014.

figure 2
Fig. 2

Data and methods

Our data comprise eyewitness accounts, sea level records from coastal tide gauges and offshore wave buoys as well as structural details of the seawall. As for methodology, we analyse eyewitness data, process and investigate sea level records through Fourier transform and conduct numerical simulations using the Flow3D-Hydro package (Flow Science 2022). Details of the data and methodology are provided in the following.

Eyewitness data

The scale of damage to the seawall and its effects led the local community to document the first-hand accounts of those most closely affected by the storms including residents, local businesses, emergency responders, politicians and engineering contractors involved in the post-storm restoration work. These records now form a permanent exhibition in the local museum in DawlishFootnote3, and some of these accounts have been transcribed into a DVD account of the disaster (Dawlish Museum 2015). We have gathered data from the Dawlish Museum, national and international news reports, social media tweets and videos. Table 1 provides a summary of the eyewitness accounts. Overall, 26 entries have been collected around the time of the incident. Our analysis of the eyewitness data is provided in the third column of Table 1 and is expanded in Sect. 3.Table 1 Eyewitness accounts of damage to the Dawlish railway due to the February 2014 storm and our interpretations

Full size table

Sea level data and wave environment

Our sea level data are a collection of three tide gauge stations (Newlyn, Devonport and Swanage Pier—Fig. 5a) owned and operated by the UK National Tide and Sea Level FacilityFootnote4 for the Environment Agency and four offshore wave buoys (Dawlish, West Bay, Torbay and Chesil Beach—Fig. 6a). The tide gauge sites are all fitted with POL-EKO (www.pol-eko.com.pl) data loggers. Newlyn has a Munro float gauge with one full tide and one mid-tide pneumatic bubbler system. Devonport has a three-channel data pneumatic bubbler system, and Swanage Pier consists of a pneumatic gauge. Each has a sampling interval of 15 min, except for Swanage Pier which has a sampling interval of 10 min. The tide gauges are located within the port areas, whereas the offshore wave buoys are situated approximately 2—3.3 km from the coast at water depths of 10–15 m. The wave buoys are all Datawell Wavemaker Mk III unitsFootnote5 and come with sampling interval of 0.78 s. The buoys have a maximum saturation amplitude of 20.5 m for recording the incident waves which implies that every wave larger than this threshold will be recorded at 20.5 m. The data are provided by the British Oceanographic Data CentreFootnote6 for tide gauges and the Channel Coastal ObservatoryFootnote7 for wave buoys.

Sea level analysis

The sea level data underwent quality control to remove outliers and spikes as well as gaps in data (e.g. Heidarzadeh et al. 2022; Heidarzadeh and Satake 2015). We processed the time series of the sea level data using the Matlab signal processing tool (MathWorks 2018). For calculations of the tidal signals, we applied the tidal package TIDALFIT (Grinsted 2008), which is based on fitting tidal harmonics to the observed sea level data. To calculate the surge signals, we applied a 30-min moving average filter to the de-tided data in order to remove all wind, swell and infra-gravity waves from the time series. Based on the surge analysis and the variations of the surge component before the time period of the incident, an error margin of approximately ± 10 cm is identified for our surge analysis. Spectral analysis of the wave buoy data is performed using the fast Fourier transform (FFT) of Matlab package (Mathworks 2018).

Numerical modelling

Numerical modelling of wave-structure interaction is conducted using the computational fluid dynamics package Flow3D-Hydro version 1.1 (Flow Science 2022). Flow3D-Hydro solves the transient Navier–Stokes equations of conservation of mass and momentum using a finite difference method and on Eulerian and Lagrangian frameworks (Flow Science 2022). The aforementioned governing equations are:

∇.u=0∇.u=0

(1)

∂u∂t+u.∇u=−∇Pρ+υ∇2u+g∂u∂t+u.∇u=−∇Pρ+υ∇2u+g

(2)

where uu is the velocity vector, PP is the pressure, ρρ is the water density, υυ is the kinematic viscosity and gg is the gravitational acceleration. A Fractional Area/Volume Obstacle Representation (FAVOR) is adapted in Flow3D-Hydro, which applies solid boundaries within the Eulerian grid and calculates the fraction of areas and volume in partially blocked volume in order to compute flows on corresponding boundaries (Hirt and Nichols 1981). We validated the numerical modelling through comparing the results with Sainflou’s analytical equation for the design of vertical seawalls (Sainflou 1928; Ackhurst 2020), which is as follows:

pd=ρgHcoshk(d+z)coshkdcosσtpd=ρgHcoshk(d+z)coshkdcosσt

(3)

where pdpd is the hydrodynamic pressure, ρρ is the water density, gg is the gravitational acceleration, HH is the wave height, dd is the water depth, kk is the wavenumber, zz is the difference in still water level and mean water level, σσ is the angular frequency and tt is the time. The Sainflou’s equation (Eq. 3) is used to calculate the dynamic pressure from wave action, which is combined with static pressure on the seawall.

Using Flow3D-Hydro, a model of the Dawlish seawall was made with a computational domain which is 250.0 m in length, 15.0 m in height and 0.375 m in width (Fig. 3a). The computational domain was discretised using a single uniform grid with a mesh size of 0.125 m. The model has a wave boundary at the left side of the domain (x-min), an outflow boundary on the right side (x-max), a symmetry boundary at the bottom (z-min) and a wall boundary at the top (z-max). A wall boundary implies that water or waves are unable to pass through the boundary, whereas a symmetry boundary means that the two edges of the boundary are identical and therefore there is no flow through it. The water is considered incompressible in our model. For volume of fluid advection for the wave boundary (i.e. the left-side boundary) in our simulations, we utilised the “Split Lagrangian Method”, which guarantees the best accuracy (Flow Science, 2022).

figure 3
Fig. 3

The stability of the numerical scheme is controlled and maintained through checking the Courant number (CC) as given in the following:

C=VΔtΔxC=VΔtΔx

(4)

where VV is the velocity of the flow, ΔtΔt is the time step and ΔxΔx is the spatial step (i.e. grid size). For stability and convergence of the numerical simulations, the Courant number must be sufficiently below one (Courant et al. 1928). This is maintained by a careful adjustment of the ΔxΔx and ΔtΔt selections. Flow3D-Hydro applies a dynamic Courant number, meaning the program adjusts the value of time step (ΔtΔt) during the simulations to achieve a balance between accuracy of results and speed of simulation. In our simulation, the time step was in the range ΔtΔt = 0.0051—0.051 s.

In order to achieve the most efficient mesh resolution, we varied cell size for five values of ΔxΔx = 0.1 m, 0.125 m, 0.15 m, 0.175 m and 0.20 m. Simulations were performed for all mesh sizes, and the results were compared in terms of convergence, stability and speed of simulation (Fig. 3). A linear wave with an amplitude of 1.5 m and a period of 6 s was used for these optimisation simulations. We considered wave time histories at two gauges A and B and recorded the waves from simulations using different mesh sizes (Fig. 3). Although the results are close (Fig. 3), some limited deviations are observed for larger mesh sizes of 0.20 m and 0.175 m. We therefore selected mesh size of 0.125 m as the optimum, giving an extra safety margin as a conservative solution.

The pressure from the incident waves on the vertical wall is validated in our model by comparing them with the analytical equation of Sainflou (1928), Eq. (3), which is one of the most common set of equations for design of coastal structures (Fig. 4). The model was tested by running a linear wave of period 6 s and wave amplitude of 1.5 m against the wall, with a still water level of 4.5 m. It can be seen that the model results are very close to those from analytical equations of Sainflou (1928), indicating that our numerical model is accurately modelling the wave-structure interaction (Fig. 4).

figure 4
Fig. 4

Eyewitness account analysis

Contemporary reporting of the 4th and 5th February 2014 storms by the main national news outlets in the UK highlights the extreme nature of the events and the significant damage and disruption they were likely to have on the communities of the south-west of England. In interviews, this was reinforced by Network Rail engineers who, even at this early stage, were forecasting remedial engineering works to last for at least 6 weeks. One week later, following subsequent storms the cascading nature of the events was obvious. Multiple breaches of the seawall had taken place with up to 35 separate landslide events and significant damage to parapet walls along the coastal route also were reported. Residents of the area reported extreme effects of the storm, one likening it to an earthquake and reporting water ingress through doors windows and even through vertical chimneys (Table 1). This suggests extreme wave overtopping volumes and large wave impact forces. One resident described the structural effects as: “the house was jumping up and down on its footings”.

Disaster management plans were quickly and effectively put into action by the local council, police service and National Rail. A major incident was declared, and decisions regarding evacuation of the residents under threat were taken around 2100 h on the night of 4th February when reports of initial damage to the seawall were received (Table 1). Local hotels were asked to provide short-term refuge to residents while local leisure facilities were prepared to accept residents later that evening. Initial repair work to the railway line was hampered by successive high spring tides and storms in the following days although significant progress was still made when weather conditions permitted (Table 1).

Sea level observations and spectral analysis

The results of surge and wave analyses are presented in Figs. 5 and 6. A surge height of up to 0.8 m was recorded in the examined tide gauge stations (Fig. 5b-d). Two main episodes of high surge heights are identified: the first surge started on 3rd February 2014 at 03:00 (UTC) and lasted until 4th of February 2014 at 00:00; the second event occurred in the period 4th February 2014 15:00 to 5th February 2014 at 17:00 (Fig. 5b-d). These data imply surge durations of 21 h and 26 h for the first and the second events, respectively. Based on the surge data in Fig. 5, we note that the storm event of early February 2014 and the associated surges was a relatively powerful one, which impacted at least 230 km of the south coast of England, from Land’s End to Weymouth, with large surge heights.

figure 5
Fig. 5
figure 6
Fig. 6

Based on wave buoy records, the maximum recorded amplitudes are at least 20.5 m in Dawlish and West Bay, 1.9 m in Tor Bay and 4.9 m in Chesil (Fig. 6a-b). The buoys at Tor Bay and Chesil recorded dual peak period bands of 4–8 and 8–12 s, whereas at Dawlish and West Bay registered triple peak period bands at 4–8, 8–12 and 20–25 s (Fig. 6c, d). It is important to note that the long-period waves at 20–25 s occur with short durations (approximately 2 min) while the waves at the other two bands of 4–8 and 8–12 s appear to be present at all times during the storm event.

The wave component at the period band of 4–8 s can be most likely attributed to normal coastal waves while the one at 8–12 s, which is longer, is most likely the swell component of the storm. Regarding the third component of the waves with long period of 20 -25 s, which occurs with short durations of 2 min, there are two hypotheses; it is either the result of a local (port and harbour) and regional (the Lyme Bay) oscillations (eg. Rabinovich 1997; Heidarzadeh and Satake 2014; Wang et al. 1992), or due to an abnormally long swell. To test the first hypothesis, we consider various water bodies such as Lyme Bay (approximate dimensions of 70 km × 20 km with an average water depth of 30 m; Fig. 6), several local bays (approximate dimensions of 3.6 km × 0.6 km with an average water depth of 6 m) and harbours (approximate dimensions of 0.5 km × 0.5 km with an average water depth of 4 m). Their water depths are based on the online Marine navigation website.Footnote8 According to Rabinovich (2010), the oscillation modes of a semi-enclosed rectangle basin are given by the following equation:

Tmn=2gd−−√[(m2L)2+(nW)2]−1/2Tmn=2gd[(m2L)2+(nW)2]−1/2

(5)

where TmnTmn is the oscillation period, gg is the gravitational acceleration, dd is the water depth, LL is the length of the basin, WW is the width of the basin, m=1,2,3,…m=1,2,3,… and n=0,1,2,3,…n=0,1,2,3,…; mm and nn are the counters of the different modes. Applying Eq. (5) to the aforementioned water bodies results in oscillation modes of at least 5 min, which is far longer than the observed period of 20–25 s. Therefore, we rule out the first hypothesis and infer that the long period of 20–25 s is most likely a long swell wave coming from distant sources. As discussed by Rabinovich (1997) and Wang et al. (2022), comparison between sea level spectra before and after the incident is a useful method to distinguish the spectrum of the weather event. A visual inspection of Fig. 6 reveals that the forcing at the period band of 20–25 s is non-existent before the incident.

Numerical simulations of wave loading and overtopping

Based on the results of sea level data analyses in the previous section (Fig. 6), we use a dual peak wave spectrum with peak periods of 10.0 s and 25.0 s for numerical simulations because such a wave would be comprised of the most energetic signals of the storm. For variations of water depth (2.0–4.0 m), coastal wave amplitude (0.5–1.5 m) (Fig. 7) and storm surge height (0.5–0.8 m) (Fig. 5), we developed 20 scenarios (Scn) which we used in numerical simulations (Table 2). Data during the incident indicated that water depth was up to the crest level of the seawall (approximately 4 m water depth); therefore, we varied water depth from 2 to 4 m in our simulation scenarios. Regarding wave amplitudes, we referred to the variations at a nearby tide gauge station (West Bay) which showed wave amplitude up to 1.2 m (Fig. 7). Therefore, wave amplitude was varied from 0.5 m to 1.5 m by considering a factor a safety of 25% for the maximum wave amplitude. As for the storm surge component, time series of storm surges calculated at three coastal stations adjacent to Dawlish showed that it was in the range of 0.5 m to 0.8 m (Fig. 5). These 20 scenarios would help to study uncertainties associated with wave amplitudes and pressures. Figure 8 shows snapshots of wave propagation and impacts on the seawall at different times.

figure 7
Fig. 7

Table 2 The 20 scenarios considered for numerical simulations in this study

Full size table

figure 8
Fig. 8

Results of wave amplitude simulations

Large wave amplitudes can induce significant wave forcing on the structure and cause overtopping of the seawall, which could eventually cascade to other hazards such as erosion of the backfill and scour (Adams and Heidarzadeh, 2021). The first 10 scenarios of our modelling efforts are for the same incident wave amplitudes of 0.5 m, which occur at different water depths (2.0–4.0 m) and storm surge heights (0.5–0.8 m) (Table 2 and Fig. 9). This is because we aim at studying the impacts of effective water depth (deff—the sum of mean sea level and surge height) on the time histories of wave amplitudes as the storm evolves. As seen in Fig. 9a, by decreasing effective water depth, wave amplitude increases. For example, for Scn-1 with effective depth of 4.5 m, the maximum amplitude of the first wave is 1.6 m, whereas it is 2.9 m for Scn-2 with effective depth of 3.5 m. However, due to intensive reflections and interferences of the waves in front of the vertical seawall, such a relationship is barely seen for the second and the third wave peaks. It is important to note that the later peaks (second or third) produce the largest waves rather than the first wave. Extraordinary wave amplifications are seen for the Scn-2 (deff = 3.5 m) and Scn-7 (deff = 3.3 m), where the corresponding wave amplitudes are 4.5 m and 3.7 m, respectively. This may indicate that the effective water depth of deff = 3.3–3.5 m is possibly a critical water depth for this structure resulting in maximum wave amplitudes under similar storms. In the second wave impact, the combined wave height (i.e. the wave amplitude plus the effective water depth), which is ultimately an indicator of wave overtopping, shows that the largest wave heights are generated by Scn-2, 7 and 8 (Fig. 9a) with effective water depths of 3.5 m, 3.3 m and 3.8 m and combined heights of 8.0 m, 7.0 m and 6.9 m (Fig. 9b). Since the height of seawall is 5.4 m, the combined wave heights for Scn-2, 7 and 8 are greater than the crest height of the seawall by 2.6 m, 1.6 m and 1.5 m, respectively, which indicates wave overtopping.

figure 9
Fig. 9

For scenarios 11–20 (Fig. 10), with incident wave amplitudes of 1.5 m (Table 2), the largest wave amplitudes are produced by Scn-17 (deff = 3.3 m), Scn-13 (deff = 2.5 m) and Scn-12 (deff = 3.5 m), which are 5.6 m, 5.1 m and 4.5 m. The maximum combined wave heights belong to Scn-11 (deff = 4.5 m) and Scn-17 (deff = 3.3 m), with combined wave heights of 9.0 m and 8.9 m (Fig. 10b), which are greater than the crest height of the seawall by 4.6 m and 3.5 m, respectively.

figure 10
Fig. 10

Our simulations for all 20 scenarios reveal that the first wave is not always the largest and wave interactions, reflections and interferences play major roles in amplifying the waves in front of the seawall. This is primarily because the wall is fully vertical and therefore has a reflection coefficient of close to one (i.e. full reflection). Simulations show that the combined wave height is up to 4.6 m higher than the crest height of the wall, implying that severe overtopping would be expected.

Results of wave loading calculations

The pressure calculations for scenarios 1–10 are given in Fig. 11 and those of scenarios 11–20 in Fig. 12. The total pressure distribution in Figs. 1112 mostly follows a triangular shape with maximum pressure at the seafloor as expected from the Sainflou (1928) design equations. These pressure plots comprise both static (due to mean sea level in front of the wall) and dynamic (combined effects of surge and wave) pressures. For incident wave amplitudes of 0.5 m (Fig. 11), the maximum wave pressure varies in the range of 35–63 kPa. At the sea surface, it is in the range of 4–20 kPa (Fig. 11). For some scenarios (Scn-2 and 7), the pressure distribution deviates from a triangular shape and shows larger pressures at the top, which is attributed to the wave impacts and partial breaking at the sea surface. This adds an additional triangle-shaped pressure distribution at the sea surface elevation consistent with the design procedure developed by Goda (2000) for braking waves. The maximum force on the seawall due to scenarios 1–10, which is calculated by integrating the maximum pressure distribution over the wave-facing surface of the seawall, is in the range of 92–190 KN (Table 2).

figure 11
Fig. 11
figure 12
Fig. 12

For scenarios 11–20, with incident wave amplitude of 1.5 m, wave pressures of 45–78 kPa and 7–120 kPa, for  the bottom and top of the wall, respectively, were observed (Fig. 12). Most of the plots show a triangular pressure distribution, except for Scn-11 and 15. A significant increase in wave impact pressure is seen for Scn-15 at the top of the structure, where a maximum pressure of approximately 120 kPa is produced while other scenarios give a pressure of 7–32 kPa for the sea surface. In other words, the pressure from Scn-15 is approximately four times larger than the other scenarios. Such a significant increase of the pressure at the top is most likely attributed to the breaking wave impact loads as detailed by Goda (2000) and Cuomo et al. (2010). The wave simulation snapshots in Fig. 8 show that the wave breaks before reaching the wall. The maximum force due to scenarios 11–20 is 120–286 KN.

The breaking wave impacts peaking at 286 KN in our simulations suggest destabilisation of the upper masonry blocks, probably by grout malfunction. This significant impact force initiated the failure of the seawall which in turn caused extensive ballast erosion. Wave impact damage was proposed by Adams and Heidarzadeh (2021) as one of the primary mechanisms in the 2014 Dawlish disaster. In the multi-hazard risk model proposed by these authors, damage mechanism III (failure pathway 5 in Adams and Heidarzadeh, 2021) was characterised by wave impact force causing damage to the masonry elements, leading to failure of the upper sections of the seawall and loss of infill material. As blocks were removed, access to the track bed was increased for inbound waves allowing infill material from behind the seawall to be fluidised and subsequently removed by backwash. The loss of infill material critically compromised the stability of the seawall and directly led to structural failure. In parallel, significant wave overtopping (discussed in the next section) led to ballast washout and cascaded, in combination with masonry damage, to catastrophic failure of the wall and suspension of the rails in mid-air (Fig. 1b), leaving the railway inoperable for two months.

Wave Overtopping

The two most important factors contributing to the 2014 Dawlish railway catastrophe were wave impact forces and overtopping. Figure 13 gives the instantaneous overtopping rates for different scenarios, which experienced overtopping. It can be seen that the overtopping rates range from 0.5 m3/s/m to 16.1 m3/s/m (Fig. 13). Time histories of the wave overtopping rates show that the phenomenon occurs intermittently, and each time lasts 1.0–7.0 s. It is clear that the longer the overtopping time, the larger the volume of the water poured on the structure. The largest wave overtopping rates of 16.1 m3/s/m and 14.4 m3/s/m belong to Scn-20 and 11, respectively. These are the two scenarios that also give the largest combined wave heights (Fig. 10b).

figure 13
Fig. 13

The cumulative overtopping curves (Figs. 1415) show the total water volume overtopped the structure during the entire simulation time. This is an important hazard factor as it determines the level of soil saturation, water pore pressure in the soil and soil erosion (Van der Meer et al. 2018). The maximum volume belongs to Scn-20, which is 65.0 m3/m (m-cubed of water per metre length of the wall). The overtopping volumes are 42.7 m3/m for Scn-11 and 28.8 m3/m for Scn-19. The overtopping volume is in the range of 0.7–65.0 m3/m for all scenarios.

figure 14
Fig. 14
figure 15
Fig. 15

For comparison, we compare our modelling results with those estimated using empirical equations. For the case of the Dawlish seawall, we apply the equation proposed by Van Der Meer et al. (2018) to estimate wave overtopping rates, based on a set of decision criteria which are the influence of foreshore, vertical wall, possible breaking waves and low freeboard:

qgH3m−−−−√=0.0155(Hmhs)12e(−2.2RcHm)qgHm3=0.0155(Hmhs)12e(−2.2RcHm)

(6)

where qq is the mean overtopping rate per metre length of the seawall (m3/s/m), gg is the acceleration due to gravity, HmHm is the incident wave height at the toe of the structure, RcRc is the wall crest height above mean sea level, hshs is the deep-water significant wave height and e(x)e(x) is the exponential function. It is noted that Eq. (6) is valid for 0.1<RcHm<1.350.1<RcHm<1.35. For the case of the Dawlish seawall and considering the scenarios with larger incident wave amplitude of 1.5 m (hshs= 1.5 m), the incident wave height at the toe of the structure is HmHm = 2.2—5.6 m, and the wall crest height above mean sea level is RcRc = 0.6–2.9 m. As a result, Eq. (6) gives mean overtopping rates up to approximately 2.9 m3/s/m. A visual inspection of simulated overtopping rates in Fig. 13 for Scn 11–20 shows that the mean value of the simulated overtopping rates (Fig. 13) is close to estimates using Eq. (6).

Discussion and conclusions

We applied a combination of eyewitness account analysis, sea level data analysis and numerical modelling in combination with our engineering judgement to explain the damage to the Dawlish railway seawall in February 2014. Main findings are:

  • Eyewitness data analysis showed that the extreme nature of the event was well forecasted in the hours prior to the storm impact; however, the magnitude of the risks to the structures was not well understood. Multiple hazards were activated simultaneously, and the effects cascaded to amplify the damage. Disaster management was effective, exemplified by the establishment of an emergency rendezvous point and temporary evacuation centre during the storm, indicating a high level of hazard awareness and preparedness.
  • Based on sea level data analysis, we identified triple peak period bands at 4–8, 8–12 and 20–25 s in the sea level data. Storm surge heights and wave oscillations were up to 0.8 m and 1.5 m, respectively.
  • Based on the numerical simulations of 20 scenarios with different water depths, incident wave amplitudes, surge heights and peak periods, we found that the wave oscillations at the foot of the seawall result in multiple wave interactions and interferences. Consequently, large wave amplitudes, up to 4.6 m higher than the height of the seawall, were generated and overtopped the wall. Extreme impulsive wave impact forces of up to 286 KN were generated by the waves interacting with the seawall.
  • We measured maximum wave overtopping rates of 0.5–16.1 m3/s/m for our scenarios. The cumulative overtopping water volumes per metre length of the wall were 0.7–65.0 m3/m.
  • Analysis of all the evidence combined with our engineering judgement suggests that the most likely initiating cause of the failure was impulsive wave impact forces destabilising one or more grouted joints between adjacent masonry blocks in the wall. Maximum observed pressures of 286 KN in our simulations are four times greater in magnitude than background pressures leading to block removal and initiating failure. Therefore, the sequence of cascading events was :1) impulsive wave impact force causing damage to masonry, 2) failure of the upper sections of the seawall, 3) loss of infill resulting in a reduction of structural strength in the landward direction, 4) ballast washout as wave overtopping and inbound wave activity increased and 5) progressive structural failure following successive tides.

From a risk mitigation point of view, the stability of the seawall in the face of future energetic cyclonic storm events and sea level rise will become a critical factor in protecting the rail network. Mitigation efforts will involve significant infrastructure investment to strengthen the civil engineering assets combined with improved hazard warning systems consisting of meteorological forecasting and real-time wave observations and instrumentation. These efforts must take into account the amenity value of coastal railway infrastructure to local communities and the significant number of tourists who visit every year. In this regard, public awareness and active engagement in the planning and execution of the project will be crucial in order to secure local stakeholder support for the significant infrastructure project that will be required for future resilience.

Notes

  1. https://www.networkrail.co.uk/..
  2. https://www.flow3d.com/products/flow-3d-hydro/.
  3. https://www.devonmuseums.net/Dawlish-Museum/Devon-Museums/.
  4. https://ntslf.org/.
  5. https://www.datawell.nl/Products/Buoys/DirectionalWaveriderMkIII.aspx.
  6. https://www.bodc.ac.uk/.
  7. https://coastalmonitoring.org/cco/.
  8. https://webapp.navionics.com/#boating@8&key=iactHlwfP.

References

Download references

Acknowledgements

We are grateful to Brunel University London for administering the scholarship awarded to KA. The Flow3D-Hydro used in this research for numerical modelling is licenced to Brunel University London through an academic programme contract. We sincerely thank Prof Harsh Gupta (Editor-in-Chief) and two anonymous reviewers for their constructive review comments.

Funding

This project was funded by the UK Engineering and Physical Sciences Research Council (EPSRC) through a PhD scholarship to Keith Adams.

Author information

Authors and Affiliations

  1. Department of Civil and Environmental Engineering, Brunel University London, Uxbridge, UB8 3PH, UKKeith Adams
  2. Department of Architecture and Civil Engineering, University of Bath, Bath, BA2 7AY, UKMohammad Heidarzadeh

Corresponding author

Correspondence to Keith Adams.

Ethics declarations

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose.

Availability of data

All data used in this study are provided in the body of the article.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Adams, K., Heidarzadeh, M. Extratropical cyclone damage to the seawall in Dawlish, UK: eyewitness accounts, sea level analysis and numerical modelling. Nat Hazards (2022). https://doi.org/10.1007/s11069-022-05692-2

Download citation

  • Received17 May 2022
  • Accepted17 October 2022
  • Published14 November 2022
  • DOIhttps://doi.org/10.1007/s11069-022-05692-2

Share this article

Anyone you share the following link with will be able to read this content:Get shareable link

Provided by the Springer Nature SharedIt content-sharing initiative

Keywords

  • Storm surge
  • Cyclone
  • Railway
  • Climate change
  • Infrastructure
  • Resilience
Propagation of Landslide Surge in Curved River Channel and Its Interaction with Dam

굽은 강둑 산사태의 팽창 전파 및 댐과의 상호 작용, 곡선하천의 산사태 해일 전파 및 댐과의 상호작용

굽은 강둑 산사태의 팽창 전파 및 댐과의 상호 작용

펑후이, 황야지에    

  1. 수자원 보존 및 환경 학교, Three Gorges University, Yichang, Hubei 443000
  • 收稿日期:2021-08-19 修回日期:2021-09-30 发布日期:2022-10-13
  • 通讯作者: Huang Yajie (1993-), Shangqiu, Henan, 석사 학위, 그의 연구 방향은 수리 구조입니다. 이메일: master_hyj@163.com
  • 作者简介:Peng Hui(1976-)는 후베이성 ​​이창에서 태어나 교수, 의사, 박사 지도교수로 주로 수력 구조의 교육 및 연구에 종사했습니다. 이메일:hpeng1976@163.com
  • 基金资助:국가핵심연구개발사업(2018YFC1508801-4)

곡선하천의 산사태 해일 전파 및 댐과의 상호작용

PENG Hui, HUANG Ya-jie    

  1. 중국 삼협대학 수자원환경대학 이창 443000 중국
  • Received:2021-08-19 Revised:2021-09-30 Published:2022-10-13

Abstract

추상적인:저수지 제방 산사태는 일반적인 지질학적 위험으로, 제때에 미리 경고하지 않으면 하천에 해일파가 발생하여 하천 교통이나 인근 수자원 보호 시설의 안전을 위험에 빠뜨릴 수 있습니다. 저수지 제방 산사태로 인한 해일파 전파 전파 Flow-3D를 이용하여 하류 댐과의 상호작용을 시뮬레이션 하였다. 수리학적 물리적 모델 시험의 타당성과 정확성을 검증하기 위하여 3차원 산사태 해지 모델을 구축하였다. 수면 높이 변화와 서지의 전파 과정에 대한 수리학적 물리적 모델 테스트. 그 동안,가장 위험한 수심과 입사각 조건은 다양한 조건에서 댐과 산사태 해일 사이의 상호 작용을 분석하여 얻었습니다. 엔지니어링 사례는 최대 동적 수두가 해일 높이의 수두보다 작고 물을 따라 감소한다는 것을 보여주었습니다. 이 경우, 서지의 정적 최대 수두에 따라 계산된 댐의 응력은 안전합니다.

As a common geological hazard,reservoir bank landslide would most probably induce surge waves in river if not prewarned in time,endangering river traffic or the safety of nearby water conservancy facilities.The propagation of surge wave induced by the landslide of curved river bank in reservoir and its interaction with downstream dam were simulated by using Flow-3D.A three-dimensional landslide surge model was constructed to verify the validity and accuracy of hydraulic physical model test.The result of the three-dimensional numerical simulation was in good agreement with that of hydraulic physical model test in terms of the water surface height change and the propagation process of the surge.In the mean time,the most dangerous water depth and incident angle conditions were obtained by analyzing the interaction between the dam and the landslide surge under different conditions.Engineering examples demonstrated that the maximum dynamic water head was smaller than the water head of surge height,and reduced along the water depth direction.In such cases,the stress of the dam calculated according to the static maximum water head of the surge is safe.

Key words

슬라이드 서지, 곡선 수로형 저수지, 수치 시뮬레이션, 동적 수압, 중력 댐, slide surges, curved channel type reservoirs, numerical simulation, dynamic water pressure, gravity dam

The failure propagation of weakly stable sediment: A reason for the formation of high-velocity turbidity currents in submarine canyons

약한 안정 퇴적물의 실패 전파: 해저 협곡에서 고속 탁도 흐름이 형성되는 이유

Abstract

Abstract해저 협곡에서 탁도의 장거리 이동은 많은 양의 퇴적물을 심해 평원으로 운반할 수 있습니다. 이전 연구에서는 5.9~28.0m/s 범위의 다중 케이블 손상 이벤트에서 파생된 탁도 전류 속도와 0.15~7.2m/s 사이의 현장 관찰 결과에서 명백한 차이가 있음을 보여줍니다. 따라서 해저 환경의 탁한 유체가 해저 협곡을 고속으로 장거리로 흐를 수 있는지에 대한 질문이 남아 있습니다. 연구실 시험의 결합을 통해 해저협곡의 탁류의 고속 및 장거리 운동을 설명하기 위해 약안정 퇴적물 기반의 새로운 모델(약안정 퇴적물에 대한 파손 전파 모델 제안, 줄여서 WSS-PFP 모델)을 제안합니다. 및 수치 아날로그. 이 모델은 두 가지 메커니즘을 기반으로 합니다. 1) 원래 탁도류는 약하게 안정한 퇴적층의 불안정화를 촉발하고 연질 퇴적물의 불안정화 및 하류 방향으로의 이동을 촉진하고 2) 원래 탁도류가 협곡으로 이동할 때 형성되는 여기파가 불안정화로 이어진다. 하류 방향으로 약하게 안정한 퇴적물의 수송. 제안된 모델은 심해 퇴적, 오염 물질 이동 및 광 케이블 손상 연구를 위한 동적 프로세스 해석을 제공할 것입니다.

The long-distance movement of turbidity currents in submarine canyons can transport large amounts of sediment to deep-sea plains. Previous studies show obvious differences in the turbidity current velocities derived from the multiple cables damage events ranging from 5.9 to 28.0 m/s and those of field observations between 0.15 and 7.2 m/s. Therefore, questions remain regarding whether a turbid fluid in an undersea environment can flow through a submarine canyon for a long distance at a high speed. A new model based on weakly stable sediment is proposed (proposed failure propagation model for weakly stable sediments, WSS-PFP model for short) to explain the high-speed and long-range motion of turbidity currents in submarine canyons through the combination of laboratory tests and numerical analogs. The model is based on two mechanisms: 1) the original turbidity current triggers the destabilization of the weakly stable sediment bed and promotes the destabilization and transport of the soft sediment in the downstream direction and 2) the excitation wave that forms when the original turbidity current moves into the canyon leads to the destabilization and transport of the weakly stable sediment in the downstream direction. The proposed model will provide dynamic process interpretation for the study of deep-sea deposition, pollutant transport, and optical cable damage.

Keyword

  • turbidity current
  • excitation wave
  • dense basal layer
  • velocity
  • WSS-PFP model

References

Download references

Acknowledgment

We thank Hanru WU from Ocean University of China for his help in thesis writing, and Hao TIAN and Chenxi WANG from Ocean University of China for their helps in the preparation of the experimental materials. Guohui XU is responsible for the development of the initial concept, processing of test data, and management of coauthor contributions to the paper; Yupeng REN for the experiment setup and drafting of the paper; Yi ZHANG and Xingbei XU for the simulation part of the experiment; Houjie WANG for writing guidance; Zhiyuan CHEN for the experiment setup.

Author information

Authors and Affiliations

  1. Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Qingdao, 266100, ChinaYupeng Ren, Yi Zhang, Guohui Xu, Xingbei Xu & Zhiyuan Chen
  2. Shandong Provincial Key Laboratory of Marine Environment and Geological Engineering, Ocean University of China, Qingdao, 266100, ChinaYupeng Ren & Houjie Wang
  3. Key Laboratory of Marine Environment and Ecology, Ocean University of China, Ministry of Education, Qingdao, 266100, ChinaYi Zhang, Guohui Xu, Xingbei Xu & Zhiyuan Chen

Corresponding author

Correspondence to Guohui Xu.

Additional information

Supported by the National Natural Science Foundation of China (Nos. 41976049, 41720104001) and the Taishan Scholar Project of Shandong Province (No. TS20190913), and the Fundamental Research Funds for the Central Universities (No. 202061028)

Data Availability Statement

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Reprints and Permissions

About this article

Cite this article

Ren, Y., Zhang, Y., Xu, G. et al. The failure propagation of weakly stable sediment: A reason for the formation of high-velocity turbidity currents in submarine canyons. J. Ocean. Limnol. (2022). https://doi.org/10.1007/s00343-022-1285-0

Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.

Numerical study of the dam-break waves and Favre waves down sloped wet rigid-bed at laboratory scale

WenjunLiuaBoWangaYakunGuobaState Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, Chengdu 610065, ChinabFaculty of Engineering & Informatics, University of Bradford, BD7 1DP, UK

Highlights

경사진 습윤층에서 댐파괴유동과 FFavre 파를 수치적으로 조사하였다.
수직 대 수평 속도의 비율이 먼저 정량화됩니다.
유동 상태는 유상 경사가 큰 후기 단계에서 크게 변경됩니다.
Favre 파도는 수직 속도와 수직 가속도에 큰 영향을 미칩니다.
베드 전단응력의 변화는 베드 기울기와 꼬리물의 영향을 받습니다.

Abstract

The bed slope and the tailwater depth are two important ones among the factors that affect the propagation of the dam-break flood and Favre waves. Most previous studies have only focused on the macroscopic characteristics of the dam-break flows or Favre waves under the condition of horizontal bed, rather than the internal movement characteristics in sloped channel. The present study applies two numerical models, namely, large eddy simulation (LES) and shallow water equations (SWEs) models embedded in the CFD software package FLOW-3D to analyze the internal movement characteristics of the dam-break flows and Favre waves, such as water level, the velocity distribution, the fluid particles acceleration and the bed shear stress, under the different bed slopes and water depth ratios. The results under the conditions considered in this study show that there is a flow state transition in the flow evolution for the steep bed slope even in water depth ratio α = 0.1 (α is the ratio of the tailwater depth to the reservoir water depth). The flow state transition shows that the wavefront changes from a breaking state to undular. Such flow transition is not observed for the horizontal slope and mild bed slope. The existence of the Favre waves leads to a significant increase of the vertical velocity and the vertical acceleration. In this situation, the SWEs model has poor prediction. Analysis reveals that the variation of the maximum bed shear stress is affected by both the bed slope and tailwater depth. Under the same bed slope (e.g., S0 = 0.02), the maximum bed shear stress position develops downstream of the dam when α = 0.1, while it develops towards the end of the reservoir when α = 0.7. For the same water depth ratio (e.g., α = 0.7), the maximum bed shear stress position always locates within the reservoir at S0 = 0.02, while it appears in the downstream of the dam for S0 = 0 and 0.003 after the flow evolves for a while. The comparison between the numerical simulation and experimental measurements shows that the LES model can predict the internal movement characteristics with satisfactory accuracy. This study improves the understanding of the effect of both the bed slope and the tailwater depth on the internal movement characteristics of the dam-break flows and Favre waves, which also provides a valuable reference for determining the flood embankment height and designing the channel bed anti-scouring facility.

Fig. 1. Sketch of related variables involved in shallow water model.
Fig. 1. Sketch of related variables involved in shallow water model.
Fig. 2. Flume model in numerical simulation.
Fig. 2. Flume model in numerical simulation.
Fig. 3. Grid sensitivity analysis (a) water surface profile; (b) velocity profile.
Fig. 3. Grid sensitivity analysis (a) water surface profile; (b) velocity profile.
Fig. 4. Sketch of experimental set-up for validating the velocity profile.
Fig. 4. Sketch of experimental set-up for validating the velocity profile.
Fig. 5. Sketch of experimental set-up for validating the bed shear stress.
Fig. 5. Sketch of experimental set-up for validating the bed shear stress.
Fig. 6. Model validation results (a) variation of the velocity profile; (b) error value of the velocity profile; (c) variation of the bed shear stress; (d) error value of the bed shear stress.
Fig. 6. Model validation results (a) variation of the velocity profile; (b) error value of the velocity profile; (c) variation of the bed shear stress; (d) error value of the bed shear stress.
Fig. 7. Schematic diagram of regional division.
Fig. 7. Schematic diagram of regional division.
Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 8. Variation of water surface profile (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 8. (continued).
Fig. 8. (continued).
Fig. 8. (continued).
Fig. 8. (continued).
Fig. 8. (continued).
Fig. 8. (continued).
Fig. 9. Froude number for α = 0.1 (a) variation with time; (b) variation with wavefront position.
Fig. 9. Froude number for α = 0.1 (a) variation with time; (b) variation with wavefront position.
Fig. 10. Characteristics of velocity distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 10. Characteristics of velocity distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 11. Average proportion of the vertical velocity (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 11. Average proportion of the vertical velocity (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 12. Bed shear stress distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 12. Bed shear stress distribution (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 12. (continued).
Fig. 12. (continued).
Fig. 13. Variation of the maximum bed shear stress position with time (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 13. Variation of the maximum bed shear stress position with time (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 14. Time when the maximum bed shear stress appears at different positions (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 14. Time when the maximum bed shear stress appears at different positions (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 15. Movement characteristics of the fluid particles (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 15. Movement characteristics of the fluid particles (a) α = 0.1; (b) α = 0.3; (c) α = 0.5; (d) α = 0.7.
Fig. 15. (continued).
Fig. 15. (continued).

Keywords

Dam-break flow, Bed slope, Wet bed, Velocity profile, Bed shear stress, Large eddy simulation

References

Barnes, M.P., Baldock, T.E. 2006. Bed shear stress measurements in dam break and swash
flows. Proceedings of International Conference on Civil and Environmental
Engineering. Hiroshima University, Japan, 28–29 September.
Biscarini, C., Francesco, S.D., Manciola, P., 2010. CFD modelling approach for dam break
flow studies. Hydrol. Earth Syst. Sc. 14, 705–718. https://doi.org/10.5194/hess-14-
705-2010.
Fig. 15. (continued).
W. Liu et al.
Journal of Hydrology 602 (2021) 126752
19
Bristeau, M.-O., Goutal, N., Sainte-Marie, J., 2011. Numerical simulations of a nonhydrostatic shallow water model. Comput. Fluids. 47 (1), 51–64. https://doi.org/
10.1016/j.compfluid.2011.02.013.
Bung, D.B., Hildebrandt, A., Oertel, M., Schlenkhoff, A., Schlurmann, T. 2008. Bore
propagation over a submerged horizontal plate by physical and numerical
simulation. Proc. 31st Intl.Conf. Coastal Eng., Hamburg, Germany, 3542–3553.
Cantero-Chinchilla, F.N., Castro-Orgaz, O., Dey, S., Ayuso, J.L., 2016. Nonhydrostatic
dam break flows. I: physical equations and numerical schemes. J. Hydraul. Eng. 142
(12), 04016068. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001205.
Castro-Orgaz, O., Chanson, H., 2020. Undular and broken surges in dam-break flows: A
review of wave breaking strategies in a boussinesq-type framework. Environ. Fluid
Mech. 154 https://doi.org/10.1007/s10652-020-09749-3.
Chang, T.-J., Kao, H.-M., Chang, K.-H., Hsu, M.-H., 2011. Numerical simulation of
shallow-water dam break flows in open channels using smoothed particle
hydrodynamics. J. Hydrol. 408 (1-2), 78–90. https://doi.org/10.1016/j.
jhydrol.2011.07.023.
Chen, H., Xu, W., Deng, J., Xue, Y., Li, J., 2009. Experimental investigation of pressure
load exerted on a downstream dam by dam-break flow. J. Hydraul. Eng. 140,
199–207. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000743.
Favre H. 1935. Etude th´eorique et exp´erimentale des ondes de translation dans les
canaux d´ecouverts. Dunod, Paris. (in French).
Flow Science Inc. 2016. Flow-3D User’s Manuals. Santa Fe NM.
Fraccarollo, L., Toro, E.F., 1995. Experimental and numerical assessment of the shallow
water model for two-dimensional dam-break type problems. J. Hydraul. Res. 33 (6),
843–864. https://doi.org/10.1080/00221689509498555.
Guo, Y., Wu, X., Pan, C., Zhang, J., 2012. Numerical simulation of the tidal flow and
suspended sediment transport in the qiantang estuary. J Waterw. Port Coastal. 138
(3), 192–202. https://doi.org/10.1061/(ASCE)WW.1943-5460.0000118.
Guo, Y., Zhang, Z., Shi, B., 2014. Numerical simulation of gravity current descending a
slope into a linearly stratified environment. J. Hydraulic Eng. 140 (12), 04014061.
https://doi.org/10.1061/(ASCE)HY.1943-7900.0000936.
Khosronejad, A., Kang, S., Flora, K., 2019. Fully coupled free-surface flow and sediment
transport modelling of flash floods in a desert stream in the mojave desert, california.
Hydrol. Process 33 (21), 2772–2791. https://doi.org/10.1002/hyp.v33.2110.1002/
hyp.13527.
Khosronejad, A., Arabi, M.G., Angelidis, D., Bagherizadeh, E., Flora, K., Farhadzadeh, A.,
2020a. A comparative study of rigid-lid and level-set methods for LES of openchannel flows: morphodynamics. Environ. Fluid Mech. 20 (1), 145–164. https://doi.
org/10.1007/s10652-019-09703-y.
Khosronejad, A., Flora, K., Zhang, Z.X., Kang, S., 2020b. Large-eddy simulation of flash
flood propagation and sediment transport in a dry-bed desert stream. Int. J.
Sediment Res. 35 (6), 576–586. https://doi.org/10.1016/j.ijsrc.2020.02.002.
Khoshkonesh, A., Nsom, B., Gohari, S., Banejad, H., 2019. A comprehensive study of dam
break over the dry and wet beds. Ocean Eng. 188, 106279.1–106279.18. https://doi.
org/10.1016/j.oceaneng.2019.106279.
Kocaman, S., Ozmen-Cagatay, H., 2012. The effect of lateral channel contraction on dam
break flows: laboratory experiment. J. Hydrol. 432–433, 145–153. https://doi.org/
10.1016/j.jhydrol.2012.02.035.
Kocaman, S., Ozmen-Cagatay, H., 2015. Investigation of dam-break induced shock waves
impact on a vertical wall. J. Hydrol. 525, 1–12. https://doi.org/10.1016/j.
jhydrol.2015.03.040.
LaRocque, L.A., Imran, J., Chaudhry, M.H., 2013a. Experimental and numerical
investigations of two-dimensional dam-break flows. J. Hydraul. Eng. 139 (6),
569–579. https://doi.org/10.1061/(ASCE)HY.1943-7900.0000705.
Larocque, L.A., Imran, J., Chaudhry, M.H., 2013b. 3D numerical simulation of partial
breach dam-break flow using the LES and k-ε turbulence models. J. Hydraul. Res. 51,
145–157. https://doi.org/10.1080/00221686.2012.734862.
Lauber, G., Hager, W.H., 1998a. Experiments to dam break wave: Horizontal channel.
J. Hydraul. Res. 36 (3), 291–307. https://doi.org/10.1080/00221689809498620.
Lauber, G., Hager, W.H., 1998b. Experiments to dam break wave: Sloping channel.
J. Hydraul. Res. 36 (5), 761–773. https://doi.org/10.1080/00221689809498601.
Leal, J.G., Ferreira, R.M., Cardoso, A.H., 2006. Dam-break wave-front celerity.
J. Hydraul. Eng. 132 (1), 69–76. https://doi.org/10.1061/(ASCE)0733-9429(2006)
132:1(69).
Liu, W., Wang, B., Guo, Y., Zhang, J., Chen, Y., 2020. Experimental investigation on the
effects of bed slope and tailwater on dam-break flows. J. Hydrol. 590, 125256.
https://doi.org/10.1016/j.jhydrol.2020.125256.
Marche, C., Beauchemin P. EL Kayloubi, A. 1995. Etude num´erique et exp´erimentale des
ondes secondaires de Favre cons´ecutives a la rupture d’un harrage. Can. J. Civil Eng.
22, 793–801, (in French). https://doi.org/10.1139/l95-089.
Marra, D., Earl, T., Ancey, C. 2011. Experimental investigations of dam break flows down
an inclined channel. 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.
Marsooli, R., Wu, W., 2014. 3-D finite-volume model of dam-break flow over uneven
beds based on vof method. Adv. Water Resour. 70, 104–117. https://doi.org/
10.1016/j.advwatres.2014.04.020.
Miller, S., Chaudhry, M.H., 1989. Dam-break flows in curved channel. J. Hydraul. Eng.
115 (11), 1465–1478. https://doi.org/10.1061/(ASCE)0733-9429(1989)115:11
(1465).
Mohapatra, P.K., Chaudhry, M.H., 2004. Numerical solution of Boussinesq equations to
simulate dam-break flows. J. Hydraul. Eng. 130 (2), 156–159. https://doi.org/
10.1061/(ASCE)0733-9429(2004)130:2(156).
Oertel, M., Bung, D.B., 2012. Initial stage of two-dimensional dam-break waves:
laboratory versus VOF. J. Hydraul. Res. 50 (1), 89–97. https://doi.org/10.1080/
00221686.2011.639981.
Ozmen-Cagatay, H., Kocaman, S., 2012. Investigation of dam-break flow over abruptly
contracting channel with trapezoidal-shaped lateral obstacles. J. Fluids Eng. 134,
081204 https://doi.org/10.1115/1.4007154.
Ozmen-Cagatay, H., Kocaman, S., Guzel, H., 2014. Investigation of dam-break flood
waves in a dry channel with a hump. J. Hydro-environ. Res. 8 (3), 304–315. https://
doi.org/10.1016/j.jher.2014.01.005.
Park, I.R., Kim, K.S., Kim, J., Van, S.H., 2012. Numerical investigation of the effects of
turbulence intensity on dam-break flows. Ocean Eng. 42, 176–187. https://doi.org/
10.1016/j.oceaneng.2012.01.005.
Peregrine, D.H., 1966. Calculations of the development of an undular bore. J. Fluid
Mech. 25 (2), 321–330. https://doi.org/10.1017/S0022112066001678.
Savic, L.j., Holly, F.M., 1993. Dam break flood waves computed by modified Godunov
method. J. Hydraul. Res. 31 (2), 187–204. https://doi.org/10.1080/
00221689309498844.
Shigematsu, T., Liu, P., Oda, K., 2004. Numerical modeling of the initial stages of dambreak waves. J. Hydraul. Res. 42 (2), 183–195. https://doi.org/10.1080/
00221686.2004.9628303.
Smagorinsky, J., 1963. General circulation experiments with the primitive equations.
Part I: the basic experiment. Mon. Weather Rev. 91, 99–164. https://doi.org/
10.1126/science.27.693.594.
Soares-Frazao, S., Zech, Y., 2002. Undular bores and secondary waves – Experiments and
hybrid finite-volume modeling. J. Hydraul. Res. 40, 33–43. https://doi.org/
10.1080/00221680209499871.
Stansby, P.K., Chegini, A., Barnes, T.C.D., 1998. The initial stages of dam-break flow.
J. Fluid Mech. 370, 203–220. https://doi.org/10.1017/022112098001918.
Treske, A., 1994. Undular bores (favre-waves) in open channels – experimental studies.
J. Hydraul. Res. 32 (3), 355–370. https://doi.org/10.1080/00221689409498738.
Wang, B., Chen, Y., Wu, C., Dong, J., Ma, X., Song, J., 2016. A semi-analytical approach
for predicting peak discharge of floods caused by embankment dam failures. Hydrol.
Process 30 (20), 3682–3691. https://doi.org/10.1002/hyp.v30.2010.1002/
hyp.10896.
Wang, B., Chen, Y., Wu, C., Peng, Y., Ma, X., Song, J., 2017. Analytical solution of dambreak flood wave propagation in a dry sloped channel with an irregular-shaped
cross-section. J. Hydro-environ. Res. 14, 93–104. https://doi.org/10.1016/j.
jher.2016.11.003.
Wang, B., Chen, Y., Wu, C., Peng, Y., Song, J., Liu, W., Liu, X., 2018. Empirical and semianalytical models for predicting peak outflows caused by embankment dam failures.
J. Hydrol. 562, 692–702. https://doi.org/10.1016/j.jhydrol.2018.05.049.
Wang, B., Zhang, J., Chen, Y., Peng, Y., Liu, X., Liu, W., 2019. Comparison of measured
dam-break flood waves in triangular and rectangular channels. J. Hydrol. 575,
690–703. https://doi.org/10.1016/j.jhydrol.2019.05.081.
Wang, B., Liu, W., Zhang, J., Chen, Y., Wu, C., Peng, Y., Wu, Z., Liu, X., Yang, S., 2020a.
Enhancement of semi-theoretical models for predicting peak discharges in breached
embankment dams. Environ. Fluid Mech. 20 (4), 885–904. https://doi.org/10.1007/
s10652-019-09730-9.
Wang, B., Chen, Y., Peng, Y., Zhang, J., Guo, Y., 2020b. Analytical solution of shallow
water equations for ideal dam-break flood along a wet bed slope. J. Hydraul. Eng.
146 (2), 06019020. https://doi.org/10.1061/(ASCE)HY.1943-7900.0001683.
Wang, B., Liu, W., Wang, W., Zhang, J., Chen, Y., Peng, Y., Liu, X., Yang, S., 2020c.
Experimental and numerical investigations of similarity for dam-break flows on wet
bed. J. Hydrol. 583, 124598. https://doi.org/10.1016/j.jhydrol.2020.124598.
Wang, B., Liu, X., Zhang, J., Guo, Y., Chen, Y., Peng, Y., Liu, W., Yang, S., Zhang, F.,
2020d. Analytical and experimental investigations of dam-break flows in triangular
channels with wet-bed conditions. J. Hydraul. Eng. 146 (10), 04020070. https://doi.
org/10.1061/(ASCE)HY.1943-7900.0001808.
Wu, W., Wang, S., 2007. One-dimensional modeling of dam-break flow over movable
beds. J. Hydraul. Eng. 133 (1), 48–58. https://doi.org/10.1061/(ASCE)0733-9429
(2007)133:1(48).
Xia, J., Lin, B., Falconer, R.A., Wang, G., 2010. Modelling dam-break flows over mobile
beds using a 2d coupled approach. Adv. Water Resour. 33 (2), 171–183. https://doi.
org/10.1016/j.advwatres.2009.11.004.
Yang, S., Yang, W., Qin, S., Li, Q., Yang, B., 2018a. Numerical study on characteristics of
dam-break wave. Ocean Eng. 159, 358–371. https://doi.org/10.1016/j.
oceaneng.2018.04.011.
Yang, S., Yang, W., Qin, S., Li, Q., 2018b. Comparative study on calculation methods of
dam-break wave. J. Hydraul. Res. 57 (5), 702–714. https://doi.org/10.1080/
00221686.2018.1494057.

Fig. 6. Experiment of waves passing through a single block of porous medium.

Generalization of a three-layer model for wave attenuation in n-block submerged porous breakwater

NadhiraKarimaaIkhaMagdalenaabIndrianaMarcelaaMohammadFaridbaFaculty of Mathematics and Natural Sciences, Bandung Institute of Technology, 40132, IndonesiabCenter for Coastal and Marine Development, Bandung Institute of Technology, Indonesia

Highlights

•A new three-layer model for n-block submerged porous breakwaters is developed.

•New analytical approach in finding the wave transmission coefficient is presented.

•A finite volume method successfully simulates the wave attenuation process.

•Porous media blocks characteristics and configuration can optimize wave reduction.

Abstract

높은 파도 진폭은 해안선에 위험한 영향을 미치고 해안 복원력을 약화시킬 수 있습니다. 그러나 다중 다공성 매체는 해양 생태계의 환경 친화적인 해안 보호 역할을 할 수 있습니다.

이 논문에서 우리는 n개의 잠긴 다공성 미디어 블록이 있는 영역에서 파동 진폭 감소를 계산하기 위해 3층 깊이 통합 방정식을 사용합니다. 수학적 모델은 파동 전달 계수를 얻기 위해 여러 행렬 방정식을 포함하는 변수 분리 방법을 사용하여 해석적으로 해결됩니다.

이 계수는 진폭 감소의 크기에 대한 정보를 제공합니다. 또한 모델을 수치적으로 풀기 위해 지그재그 유한 체적 방법이 적용됩니다.

수치 시뮬레이션을 통해 다공성 매질 블록의 구성과 특성이 투과파 진폭을 줄이는 데 중요하다는 결론을 내렸습니다.

High wave amplitudes may cause dangerous effects on the shoreline and weaken coastal resilience. However, multiple porous media can act as environmental friendly coastal protectors of the marine ecosystem. In this paper, we use three-layer depth-integrated equations to calculate wave amplitude reduction in a domain with n submerged porous media blocks. The mathematical model is solved analytically using the separation of variables method involving several matrix equations to obtain the wave transmission coefficient. This coefficient provides information about the magnitude of amplitude reduction. Additionally, a staggered finite volume method is applied to solve the model numerically. By conducting numerical simulations, we conclude that porous media blocks’ configuration and characteristics are crucial in reducing transmitted wave amplitude.

Keywords

Three-layer equations, Submerged porous media, Wave transmission coefficient, Finite volume method

Fig. 1. Sketch of the problem configuration.
Fig. 1. Sketch of the problem configuration.
Fig. 6. Experiment of waves passing through a single block of porous medium.
Fig. 6. Experiment of waves passing through a single block of porous medium.

References

[1]M. Beck, G. Lange, Managing Coasts with Natural Solutions: Guidelines for Measuring and Valuing the Coastal Protection Services of Mangroves and Coral Reefs.

Google Scholar[2]

Y. Zhao, Y. Liu, H. Li, A. Chang

Oblique wave motion over multiple submerged porous bars near a vertical wall

J. Ocean Univ. China, 16 (2017), pp. 568-574, 10.1007/s11802-017-3333-5 View PDF

View Record in ScopusGoogle Scholar[3]C. K. Sollitt, R. H. Cross, Wave transmission through permeable breakwaters, Coast. Eng..

Google Scholar[4]J.-F. Lee, L.-F. Tu, C.-C. Liu, Nonlinear wave evolution above rectangular submerged structures, J. Mar. Sci. Technol. 22. doi:10.6119/JMST-013-0503-3.

Google Scholar[5]

Y.T. Wu, C.L. Yeh, S.-C. Hsiao

Three-dimensional numerical simulation on the interaction of solitary waves and porous breakwaters

Coast. Eng., 85 (2014), pp. 12-29

ArticleDownload PDFView Record in ScopusGoogle Scholar[6]

Y. feng Xu, X. he Xia, J. hua Wang, J. jian Chen

Numerical analysis on cnoidal wave induced response of porous seabed with definite thickness

J. Shanghai Jiao Tong Univ. (Sci.), 18 (2013), pp. 650-654, 10.1007/s12204-013-1446-6 View PDF

Google Scholar[7]

D.M. Pérez-Romero, M. Ortega-Sánchez, A. Moñino, M.A. Losada

Characteristic friction coefficient and scale effects in oscillatory porous flow

Coast. Eng., 56 (9) (2009), pp. 931-939, 10.1016/j.coastaleng.2009.05.002

ArticleDownload PDFView Record in ScopusGoogle Scholar[8]

A. Torres-Freyermuth, M. Brocchini, S. Corvaro, J.C. Pintado-Patiño

Wave attenuation over porous seabeds: a numerical study

Ocean Model., 117 (2017), pp. 28-40, 10.1016/j.ocemod.2017.07.004

ArticleDownload PDFView Record in ScopusGoogle Scholar[9]F. Hajivalie, S. M. Mahmoudof, Experimental study of energy dissipation at rectangular submerged breakwater, Proceedings of the 8th International Conference on Fluid Mechanics.

Google Scholar[10]G. T. Klonaris, A. S. Metallinos, C. D. Memos, K. A. Galani, Experimental and numerical investigation of bed morphology in the lee of porous submerged breakwaters, Coast. Eng. 155.

Google Scholar[11]

A. Kubowicz-Grajewska

Experimental investigation into wave interaction with a rubble-mound submerged breakwater (case study)

J. Mar. Sci. Technol., 22 (2) (2017), pp. 313-326 View PDF

CrossRefView Record in ScopusGoogle Scholar[12]

S.M. Mahmoudof, F. Hajivalie

Experimental study of hydraulic response of smooth submerged breakwaters to irregular waves

Oceanologia, 63 (4) (2021), pp. 448-462

ArticleDownload PDFView Record in ScopusGoogle Scholar[13]

C. Tsai, H. Chen, F. Lee

Wave transformation over submerged permeable breakwater on porous bottom

Ocean Eng., 33 (2006), pp. 1623-1643, 10.1016/j.oceaneng.2005.09.006

ArticleDownload PDFView Record in ScopusGoogle Scholar[14]

S. Rojanakamthorn, M. Isobe, A. Watanabe

A mathematical model of wave transformation over a submerged breakwater

Coastal Engineering in Japan, 31 (1989), pp. 209-234, 10.1080/05785634.1989.11924515 View PDF

View Record in ScopusGoogle Scholar[15]

Q. Lin, Q.r. Meng, D.q. Lu

Waves propagating over a two-layer porous barrier on a seabed

J. Hydrodyn., 30 (3) (2018), pp. 453-462 View PDF

CrossRefView Record in ScopusGoogle Scholar[16]X. Yu, A. T. Chwang, Wave motion through porous structures, J. Eng. Mech. 120. doi:10.1061/(ASCE)0733-9399(1994)120:5(989).

Google Scholar[17]

K.G. Vijay, V. Venkateswarlu, D. Karmakar

Scattering of gravity waves by multiple submerged rubble-mound breakwaters

Arabian J. Sci. Eng., 45 (10) (2020), pp. 8529-8550 View PDF

CrossRefView Record in ScopusGoogle Scholar[18]

I. Magdalena, G. Jonathan

Water waves resonance and its interaction with submerged breakwater

Results in Engineering, 13 (2022), Article 100343, 10.1016/j.rineng.2022.100343

ArticleDownload PDFView Record in ScopusGoogle Scholar[19]

I. Magdalena, K. Firdaus, D. Jayadi

Analytical and numerical studies for wave generated by submarine landslide

Alex. Eng. J., 61 (9) (2022), pp. 7303-7313, 10.1016/j.aej.2021.12.069

ArticleDownload PDFView Record in ScopusGoogle Scholar[20]

L. Arpaia, M. Ricchiuto, A.G. Filippini, R. Pedreros

An efficient covariant frame for the spherical shallow water equations: well balanced dg approximation and application to tsunami and storm surge

Ocean Model., 169 (2022), Article 101915, 10.1016/j.ocemod.2021.101915

ArticleDownload PDFView Record in ScopusGoogle Scholar[21]

M. Briani, G. Puppo, M. Ribot

Angle dependence in coupling conditions for shallow water equations at channel junctions

Comput. Math. Appl., 108 (2022), pp. 49-65, 10.1016/j.camwa.2021.12.021

ArticleDownload PDFView Record in ScopusGoogle Scholar[22]

I. Magdalena, G.R. Andadari, D.E. Reeve

An integrated study of wave attenuation by vegetation

Wave Motion, 110 (2022), Article 102878, 10.1016/j.wavemoti.2021.102878

ArticleDownload PDFView Record in ScopusGoogle Scholar[23]

I. Magdalena, R. La’lang, R. Mendoza

Quantification of wave attenuation in mangroves in manila bay using nonlinear shallow water equations

Results in Applied Mathematics, 12 (2021), Article 100191, 10.1016/j.rinam.2021.100191

ArticleDownload PDFView Record in ScopusGoogle Scholar[24]

K.T. Mandli

A numerical method for the two layer shallow water equations with dry states

Ocean Model., 72 (2013), pp. 80-91, 10.1016/j.ocemod.2013.08.001

ArticleDownload PDFView Record in ScopusGoogle Scholar[25]

M. Farhan, Z. Omar, F. Mebarek-Oudina, J. Raza, Z. Shah, R.V. Choudhari, O.D. Makinde

Implementation of the one-step one-hybrid block method on the nonlinear equation of a circular sector oscillator

Comput. Math. Model., 31 (2020), pp. 116-132, 10.1007/s10598-020-09480-0 View PDF

View Record in ScopusGoogle Scholar[26]

R. Djebali, F. Mebarek-Oudina, C. Rajashekhar

Similarity solution analysis of dynamic and thermal boundary layers: further formulation along a vertical flat plate

Phys. Scripta, 96 (8) (2021), Article 085206, 10.1088/1402-4896/abfe31 View PDF

View Record in ScopusGoogle Scholar[27]

M. Alkasassbeh, O. Zurni, F. Mebarek-Oudina, J. Raza

Heat transfer study of convective fin with temperature,Äêdependent internal heat generation by hybrid block method

Heat Tran. Asian Res., 48 (2019), pp. 1225-1244, 10.1002/htj.21428 View PDF

View Record in ScopusGoogle Scholar[28]

I. Magdalena, M.F. Eka Pebriansyah

Numerical treatment of finite difference method for solving dam break model on a wet-dry bed with an obstacle

Results in Engineering, 14 (2022), Article 100382, 10.1016/j.rineng.2022.100382

ArticleDownload PDFView Record in ScopusGoogle Scholar[29]

M. Uddin, S. Rasel, J.K. Adewole, K.S. Al Kalbani

Finite element simulation on the convective double diffusive water-based copper oxide nanofluid flow in a square cavity having vertical wavy surfaces in presence of hydro-magnetic field

Results in Engineering, 13 (2022), Article 100364, 10.1016/j.rineng.2022.100364

ArticleDownload PDFView Record in ScopusGoogle Scholar[30]

E.H.H. Al-Qadami, A.S. Abdurrasheed, Z. Mustaffa, K.W. Yusof, M. Malek, A.A. Ghani

Numerical modelling of flow characteristics over sharp crested triangular hump

Results in Engineering, 4 (2019), Article 100052, 10.1016/j.rineng.2019.100052

ArticleDownload PDFView Record in ScopusGoogle Scholar[31]

I. Magdalena, V. Kusnowo, M.I. Azis

Widowati, 1d-2d numerical model for wave attenuation by mangroves as a porous structure

Computation, 9 (6) (2021), pp. 1-21

Google Scholar[32]

I. Magdalena, M.F. Atras, L. Sembiring, M.A. Nugroho, R.S.B. Labay, M.P. Roque

Wave transmission by rectangular submerged breakwaters

Computation, 8 (2) (2020), pp. 1-18 View PDF

View Record in ScopusGoogle Scholar[33]

I. Magdalena, S.R. Pudjaprasetya

Numerical modeling for gravity waves over submerged porous media

Australian Journal of Basic and Applied Sciences, 9 (28) (2015), pp. 124-130

View Record in ScopusGoogle Scholar[34]

I. Magdalena, A. Hariz, M. Farid, M.S.B. Kusuma

Numerical studies using staggered finite volume for dam break flow with an obstacle through different geometries

Results in Applied Mathematics, 12 (2021), Article 100193, 10.1016/j.rinam.2021.100193

ArticleDownload PDFView Record in ScopusGoogle Scholar[35]

R. Walters, E. Hanert, J. Pietrzak, D. Le Roux

Comparison of unstructured, staggered grid methods for the shallow water equations

Ocean Model., 28 (1) (2009), pp. 106-117, 10.1016/j.ocemod.2008.12.004

the Sixth International Workshop on Unstructured Mesh Numerical Modelling of Coastal, Shelf and Ocean Flows

ArticleDownload PDFView Record in ScopusGoogle Scholar[36]

F. Mebarek-Oudina

Numerical modeling of the hydrodynamic stability in vertical annulus with heat source of different lengths, Engineering Science and Technology

Int. J., 20 (4) (2017), pp. 1324-1333, 10.1016/j.jestch.2017.08.003

ArticleDownload PDFView Record in ScopusGoogle Scholar[37]

S. Pudjaprasetya, I. Magdalena

Numerical modeling for gravity waves over submerged porous media

Australian Journal of Basic and Applied Sciences, 9 (2015), pp. 124-130

Google Scholar

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

Effect of roughness on separation zone dimensions.

Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes

조도 계수 및 역전 수준 변화가 개선된 90도 측면 분출구에서의 유동에 대한 실험적 및 수치적 연구

Maryam BagheriSeyed M. Ali ZomorodianMasih ZolghadrH. Md. AzamathullaC. Venkata Siva Rama Prasad

Abstract

측면 분기기(흡입구)의 상류 측에서 흐름 분리는 분기기 입구에서 와류를 일으키는 중요한 문제입니다. 이는 흐름의 유효 폭, 출력 용량 및 효율성을 감소시킵니다. 따라서 분리지대의 크기를 파악하고 크기를 줄이기 위한 방안을 제시하는 것이 필수적이다. 본 연구에서는 분리 구역의 치수를 줄이기 위한 방법으로 7가지 유형의 거칠기 요소를 분기구 입구에 설치하고 4가지 다른 배출(총 84번의 실험을 수행)과 함께 3개의 서로 다른 베드 반전 레벨을 조사했습니다. 또한 3D CFD(Computational Fluid Dynamics) 모델을 사용하여 분리 영역의 흐름 패턴과 치수를 평가했습니다. 결과는 거칠기 계수를 향상시키면 분리 영역 치수를 최대 38%까지 줄일 수 있는 반면, 드롭 구현 효과는 사용된 거칠기 계수를 기반으로 이 영역을 다르게 축소할 수 있음을 보여주었습니다. 두 가지 방법을 결합하면 분리 영역 치수를 최대 63%까지 줄일 수 있습니다.

Flow separation at the upstream side of lateral turnouts (intakes) is a critical issue causing eddy currents at the turnout entrance. It reduces the effective width of flow, turnout capacity and efficiency. Therefore, it is essential to identify the dimensions of the separation zone and propose remedies to reduce its dimensions. Installation of 7 types of roughening elements at the turnout entrance and 3 different bed invert levels, with 4 different discharges (making a total of 84 experiments) were examined in this study as a method to reduce the dimensions of the separation zone. Additionally, a 3-D Computational Fluid Dynamic (CFD) model was utilized to evaluate the flow pattern and dimensions of the separation zone. Results showed that enhancing the roughness coefficient can reduce the separation zone dimensions up to 38% while the drop implementation effect can scale down this area differently based on the roughness coefficient used. Combining both methods can reduce the separation zone dimensions up to 63%.

HIGHLIGHTS

Listen

  • Flow separation at the upstream side of lateral turnouts (intakes) is a critical issue causing eddy currents at the turnout entrance.
  • Installation of 7 types of roughening elements at the turnout entrance and 3 different bed level inverts were investigated.
  • Additionally, a 3-D Computational Fluid Dynamic (CFD) model was utilized to evaluate the flow.
  • Combining both methods can reduce the separation zone dimensions by up to 63%.
Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes
Experimental and numerical study of flow at a 90 degree lateral turnout with enhanced roughness coefficient and invert level changes

Keywords

discharge ratioflow separation zoneintakethree dimensional simulation

INTRODUCTION

Listen

Turnouts or intakes are amongst the oldest and most widely used hydraulic structures in irrigation networks. Turnouts are also used in water distribution, transmission networks, power generation facilities, and waste water treatment plants etc. The flows that enter a turnout have a strong momentum in the direction of the main waterway and that is why flow separation occurs inside the turnout. The horizontal vortex formed in the separation area is a suitable place for accumulation and deposition of sediments. The separation zone is a vulnerable area for sedimentation and for reduction of effective flow due to a contracted flow region in the lateral channel. Sedimentaion in the entrance of the intake can gradually be transfered into the lateral channel and decrease the capacity of the higher order channels over time (Jalili et al. 2011). On the other hand, the existence of coarse-grained materials causes erosion and destruction of the waterway side walls and bottom. In addition, sedimentation creates conditions for vegetation to take root and damage the waterway cover, which causes water to leak from its perimeter. Therefore, it is important to investigate the pattern of the flow separation area in turnouts and provide solutions to reduce the dimensions of this area.

The three-dimensional flow structure at turnouts is quite complex. In an experimental study by Neary & Odgaard (1993) in a 90-degree water turnout it was found that the secondary currents and separation zone varies from the bed to the water surface. They also found that at a 90-degree water turnout, the bed roughness and discharge ratio play a critical role in flow structure. They asserted that an explanation of sediment behavior at a diversion entrance requires a comprehensive understanding of 3D flow patterns around the lateral-channel entrance. In addition, they suggested that there is a strong similarity between flow in a channel bend and a diversion channel, and that this similarity can rationalize the use of bend flow models for estimation of 3D flow structures in diversion channels.

Some of the distinctive characteristics of dividing flow in a turnout include a zone of separation immediately near the entrance of the lateral turnout (separation zone), a contracted flow region in the branch channel (contracted flow), and a stagnation point near the downstream corner of the junction (stagnation zone). In the region downstream of the junction, along the continuous far wall, separation due to flow expansion may occur (Ramamurthy et al. 2007), that is, a separation zone. This can both reduce the turnout efficiency and the effective width of flow while increasing the sediment deposition in the turnout entrance (Jalili et al. 2011). Installation of submerged vanes in the turnout entrance is a method which is already applied to reduce the size of flow separation zones. The separation zone draws sediments and floating materials into themselves. This reduces effective cross-section area and reduces transmission capacity. These results have also been obtained in past studies, including by Ramamurthy et al. (2007) and in Jalili et al. (2011). Submerged vanes (Iowa vanes) are designed in order to modify the near-bed flow pattern and bed-sediment motion in the transverse direction of the river. The vanes are installed vertically on the channel bed, at an angle of attack which is usually oriented at 10–25 degrees to the local primary flow direction. Vane height is typically 0.2–0.5 times the local water depth during design flow conditions and vane length is 2–3 times its height (Odgaard & Wang 1991). They are vortex-generating devices that generate secondary circulation, thereby redistributing sediment within the channel cross section. Several factors affect the flow separation zone such as the ratio of lateral turnout discharge to main channel discharge, angle of lateral channel with respect to the main channel flow direction and size of applied submerged vanes. Nakato et al. (1990) found that sediment management using submerged vanes in the turnout entrance to Station 3 of the Council Bluffs plant, located on the Missouri River, is applicable and efficient. The results show submerged vanes are an appropriate solution for reduction of sediment deposition in a turnout entrance. The flow was treated as 3D and tests results were obtained for the flow characteristics of dividing flows in a 90-degree sharp-edged, junction. The main and lateral channel were rectangular with the same dimensions (Ramamurthy et al., 2007).

Keshavarzi & Habibi (2005) carried out experiments on intake with angles of 45, 67, 79 and 90 degrees in different discharge ratios and reported the optimum angle for inlet flow with the lowest flow separation area to be about 55 degrees. The predicted flow characteristics were validated using experimental data. The results indicated that the width and length of the separation zone increases with the increase in the discharge ratio Qr (ratio of outflow per unit width in the turnout to inflow per unit width in the main channel).

Abbasi et al. (2004) performed experiments to investigate the dimensions of the flow separation zone at a lateral turnout entrance. They demonstrated that the length and width of the separation zone decreases with the increasing ratio of lateral turn-out discharge. They also found that with a reducing angle of lateral turnout, the length of the separation zone scales up and width of separation zone reduces. Then they compared their observations with results of Kasthuri & Pundarikanthan (1987) who conducted some experiments in an open-channel junction formed by channels of equal width and an angle of lateral 90 degree turnout, which showed the dimensions of the separation zone in their experiments to be smaller than in previous studies. Kasthuri & Pundarikanthan (1987) studied vortex and flow separation dimensions at the entrance of a 90 degree channel. Results showed that increasing the diversion discharge ratio can reduce the length and width of the vortex area. They also showed that the length and width of the vortex area remain constant at diversion ratios greater than 0.7. Karami Moghaddam & Keshavarzi (2007) analyzed the flow characteristics in turnouts with angles of 55 and 90 degrees. They reported that the dimensions of the separation zone decrease by increasing the discharge ratio and reducing the turnout angle with respect to the main channel. Studies about flow separation zone can be found in Jalili et al. (2011)Nikbin & Borghei (2011)Seyedian et al. (2008).

Jamshidi et al. (2016) measured the dimensions of a flow separation zone in the presence of submerged vanes with five arrangements (parallel, stagger, compound, piney and butterflies). Results showed that the ratio of the width to the length of the separation zone (shape index) was between 0.2 and 0.28 for all arrangements.

Karami et al. (2017) developed a 3D computational fluid dynamic (CFD) code which was calibrated by measured data. They used the model to evaluate flow pattern, diversion ratio of discharge, strength of the secondary flow, and dimensions of the vortex inside the channel in various dikes and submerged vane installation scenarios. Results showed that the diversion ratio of discharge in the diversion channel is dependent on the width of the flow separation area in the main channel. A dike, perpendicular to the flow, doubles the ratio of diverted discharge and reduces the suspended sediment load compared with the base-line situation by creating outer arch conditions. In addition, increasing the longitudinal distance between vanes increases the velocity gradient between the vanes and leads to a more severe erosion of the bed near the vanes.Figure 1VIEW LARGEDOWNLOAD SLIDE

Laboratory channel dimensions.

Al-Zubaidy & Hilo (2021) used the Navier–Stokes equation to study the flow of incompressible fluids. Using the CFD software ANSYS Fluent 19.2, 3D flow patterns were simulated at a diversion channel. Their results showed good agreement using the comparison between the experimental and numerical results when the k-omega turbulence viscous model was employed. Simulation of the flow pattern was then done at the lateral channel junction using a variety of geometry designs. These improvements included changing the intake’s inclination angle and chamfering and rounding the inner corner of the intake mouth instead of the sharp edge. Flow parameters at the diversion including velocity streamlines, bed shear stress, and separation zone dimensions were computed in their study. The findings demonstrated that changing the 90° lateral intake geometry can improve the flow pattern and bed shear stress at the intake junction. Consequently, sedimentation and erosion problems are reduced. According to the conclusions of their study, a branching angle of 30° to 45° is the best configuration for increasing branching channel discharge, lowering branching channel sediment concentration.

The review of the literature shows that most of the studies deal with turnout angle, discharge ratio and implementation of vanes as techniques to reduce the area of the separation zone. This study examines the effect of roughness coefficient and drop implementation at the entrance of a 90-degree lateral turnout on the dimensions of the separation zone. As far as the authors are aware, these two variables have never been studied as a remedy to decrease the separation zone dimensions whilst enhancing turnout efficiency. Additionally, a three-dimensional numerical model is applied to simulate the flow pattern around the turnout. The numerical results are verified against experimental data.

METHOD

Experimental setup

Listen

The experiments were conducted in a 90 degree dividing flow laboratory channel. The main channel is 15 m long, 0.5 m wide and 0.4 m high and the branch channel is 3 m long, 0.35 m wide and 0.4 m high, as shown in Figure 1. The tests were carried out at 9.65 m from the beginning of the flume and were far enough from the inlet, so we were sure that the flow was fully developed. According to Kirkgöz & Ardiçlioğlu (1997) the length of the developing region would be approximantly 65 and 72 times the flow depth. In this study, the depth is 9 cm, which makes this condition.

Both the main and lateral channel had a slope of 0.0003 with side walls of concrete. A 100 hp pump discharged the water into a stilling basin at the entrance of the main flume. The discharge was measured using an ultrasonic discharge meter around the discharge pipe. Eighty-four experiments in total were carried out at range of 0.1<Fr<0.4 (Froude numbers in main channel and upstream of turnout). The depth of water in the main channel in the experiments was 9 cm, in which case the effect of surface tension can be considered; according to research by Zolghadr & Shafai Bejestan (2020) and Zolghadr et al. (2021), when the water depth is more than 6 cm, the effect of surface tension is reduced and can be ignored given that the separation phenomenon occurs in the boundary layer, the height of the roughness creates disturbances in growth and development of the boundary layer and, as a result, separation growth is also faced with disruption and its dimensions grow less compared to smooth surfaces. Similar conditions occur in case of drop implementation. A disturbance occurs in the growth of the boundary layer and as a result the separation zone dimensions decrease. In order to investigate the effect of roughness coefficient and drop implementation on the separation zone dimensions, four different discharges (16, 18, 21, 23 l/s) in subcritical conditions, seven Manning (Strickler) roughness coefficients (0.009, 0.011, 0.017, 0.023, 0.028, 0.030, 0.032) as shown in Figure 2 and three invert elevation differences between the main channel and lateral turnout invert (0, 5 and 10 cm) at the entrance of the turnout were considered. The Manning roughness coefficient values were selected based on available and feasible values for real conditions, so that 0.009 is equivalent to galvanized sheet roughness and selected for the baseline tests. 0.011 is for concrete with neat surface, 0.017 and 0.023 are for unfinished and gunite concrete respectively. 0.030 and 0.032 values are for concrete on irregular excavated rock (Chow 1959). The roughness coefficients were created by gluing sediment particles on a thin galvanized sheet which was installed at the upstream side of the lateral turnout. The values of roughness coefficients were calculated based on the Manning-Strickler formula. For this purpose, some uniformly graded sediment samples were prepared and the Manning roughness coefficient of each sample was determined with respect to the median size (D50) value pasted into the Manning-Strickler formula. Some KMnO4 was sifted in the main channel upstream to visualize and measure the dimensions of the separation zone. Consequently, when KMnO4 approached the lateral turnout a photo of the separation zone was taken from a top view. All the experiments were recorded and several photos were taken during the experiment after stablishment of steady flow conditions. The photos were then imported to AutoCAD to measure the separation zone dimensions. Because all the shooting was done with a high-definition camera and it was possible to zoom in, the results are very accurate.Figure 2VIEW LARGEDOWNLOAD SLIDE

Roughness plates.

The velocity values were also recorded by a one-dimensional velocity meter at 15 cm distance from the turnout entrance and in transverse direction (perpendicular to the flow direction).

The water level was also measured by depth gauges with a accuracy of 0.1 mm, and velocity in one direction with a single-dimensional KENEK LP 1100 with an accuracy of ±0.02 m/s (0–1 m/s), ± 0.04 m/s (1–2 m/s), ± 0.08 m/s (2–4 m/s), ±0.10 m/s (4–5 m/s).

Numerical simulation

ListenA FLOW-3D numerical model was utilized as a solver of the Navier-Stokes equation to simulate the three-dimensional flow field at the entrance of the turnout. The governing equations included continuity momentum equations. The continuity equation, regardless of the density of the fluid in the form of Cartesian coordinates x, y, and z, is as follows:

formula

(1)where uv, and w represent the velocity components in the x, y, and z directions, respectively; AxAy, and Az are the surface flow fractions in the xy, and z directions, respectively; VF denotes flow volume fraction; r is the density of the fluid; t is time; and Rsor refers to the source of the mass. Equations (2)–(4) show momentum equations in xy and z dimensions respectively :

formula

(2)

formula

(3)

formula

(4)where GxGy, and Gz are the accelerations caused by gravity in the xy, and z directions, respectively; and fxfy, and fz are the accelerations caused by viscosity in the xy, and z directions, respectively.

The turbulence models used in this study were the renormalized group (RNG) models. Evaluation of the concordance of the mentioned models with experimental studies showed that the RNG model provides more accurate results.

Two blocks of mesh were used to simulate the main channels and lateral turnout. The meshes were denser in the vicinity of the entrance of the turnout in order to increase the accuracy of computations. Boundary conditions for the main mesh block included inflow for the channel entrance (volumetric flow rate), outflow for the channel exit, ‘wall’ for the bed and the right boundary and ‘symmetry’ for the top (free surface) and left boundaries (turnout). The side wall roughness coefficient was given to the software as the Manning number in surface roughness of any component. Considering the restrictions in the available processor, a main mesh block with appropriate mesh size was defined to simulate the main flow field in the channel, while the nested mesh-block technique was utilized to create a very dense solution field near the roughness plate in order to provide accurate results around the plates and near the entrance of the lateral turnout. This technique reduced the number of required mesh elements by up to 60% in comparison with the method in which the mesh size of the main solution field was decreased to the required extent.

The numerical outputs are verified against experimental data. The hydraulic characteristics of the experiment are shown in Table 1.Table 1

Hydraulic conditions of the flow

Q(L/s)FrY1 (m)Q2/Q1
16 0.449 0.09 0.22 
18 0.335 0.09 0.61 
21 0.242 0.09 0.71 
23 0.180 0.09 1.04 

RESULTS AND DISCUSSION

Experimental results

Listen

During the experiments, the dimensions of the separation zone were recorded with an HD camera. Some photos were imported to AutoCad software. Then, the separation zones dimensions were measured and compared in different scenarios.

At the beginning, the flow pattern in the separation zone for four different hydraulic conditions was studied for seven different Manning roughness coefficients from 0.009 to 0.032. To compare the obtained results, roughness of 0.009 was considered as the base line. The percentage of reduction in separation zone area in different roughness coefficients is shown in Figure 3. According to this figure, by increasing the roughness of the turnout side wall, the separation zone area ratio reduces (ratio of separation zone area to turnout area). In other words, in any desired Froud number, the highest dimensions of the separation zone area are related to the lowest roughness coefficients. In Figure 3, ‘A’ is the area of the separation zone and ‘Ai’ represents the total area of the turnout.Figure 3VIEW LARGEDOWNLOAD SLIDE

Effect of roughness on separation zone dimensions.Figure 4VIEW LARGEDOWNLOAD SLIDE

Effect of roughness on separation zone dimensions.

It should be mentioned that the separation zone dimensions change with depth, so that the area is larger at the surface than near the bed. This study measured the dimensions of this area at the surface. Figure 4 show exactly where the roughness elements were located.Figure 5VIEW LARGEDOWNLOAD SLIDE

Comparison of separation zone for n=0.023 and n=0.032.

Figure 5 shows images of the separation zone at n=0.023 and n=0.032 as examples, and show that the separation area at n=0.032 is smaller than that of n=0.023.

The difference between the effect of the two 0.032 and 0.030 roughnesses is minor. In other words, the dimensions of the separation zone decreased by increasing roughness up to 0.030 and then remained with negligable changes.

In the next step, the effect of intake invert relative to the main stream (drop) on the dimensions of the separation zone was investigated. To do this, three different invert levels were considered: (1) without drop; (2) a 5 cm drop between the main canal and intake canal; and (3) a 10 cm drop between the main canal and intake canal. The without drop mode was considered as the control state. Figure 6 shows the effect of drop implementation on separation zone dimensions. Tables 2 and 3 show the reduced percentage of separation zone areas in 5 and 10 cm drop compared to no drop conditions as the base line. It was found that the best results were obtained when a 10 cm drop was implemented.Table 2

Decrease percentage of separation zone area in 5 cm drop

Frn=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
0.08 10.56 11.06 25.27 33.03 35.57 36.5 
0.121 7.66 11.14 11.88 15.93 34.59 36.25 
0.353 1.38 2.63 8.17 14.39 31.20 31.29 
0.362 11.54 19.56 25.73 37.89 38.31 

Table 3

Decrease percentage of separation zone area in 10 cm drop

Frn=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
0.047 4.30 8.75 23.47 31.22 34.96 35.13 
0.119 11.01 13.16 15.02 21.48 39.45 40.68 
0.348 3.89 5.71 9.82 16.09 29 30.96 
0.354 2.84 10.44 18.42 25.45 35.68 35.76 

Figure 6VIEW LARGEDOWNLOAD SLIDE

Effect of drop implementation on separation zone dimensions.

The combined effect of drop and roughness is shown in Figure 7. According to this figure, by installing a drop structure at the entrance of the intake, the dimensions of the separation zone scales down in any desired roughness coefficient. Results indicated that by increasing the roughness coefficient or drop implementation individually, the separation zone area decreases up to 38 and 25% respectively. However, employing both techniques simultaneously can reduce the separation zone area up to 63% (Table 4). The reason for the reduction of the dimensions of the separation zone area by drop implementation can be attributed to the increase of discharge ratio. This reduces the dimensions of the separation zone area.Table 4

Reduction in percentage of combined effect of roughness and 10 cm drop

Qin=0.011n=0.017n=0.023n=0.028n=0.030n=0.032
16 32.3 35.07 37.2 45.7 58.01 59.1 
18 44.5 34.15 36.18 48.13 54.2 56.18 
21 43.18 32.33 42.30 37.79 57.16 63.2 
23 40.56 34.5 34.09 46.25 50.12 57.2 

Figure 7VIEW LARGEDOWNLOAD SLIDE

Combined effect of roughness and drop on separation zone dimensions.

This method increases the discharge ratio (ratio of turnout to main channel discharge). The results are compatible with the literature. Some other researchers reported that increasing the discharge ratio can scale down the separation zone dimensions (Karami Moghaddam & Keshavarzi 2007Ramamurthy et al. 2007). However, these researchers employed other methods to enhance the discharge ratio. Drop implementation is simple and applicable in practice, since there is normally an elevation difference between the main and lateral canal in irrigation networks to ensure gravity flow occurance.

Table 4 depicts the decrease in percentage of the separation zone compared to base line conditions in different arrangements of the combined tests.Figure 8VIEW LARGEDOWNLOAD SLIDE

Velocity profiles for various roughness coefficients along turnout width.

A comparison between the proposed methods introduced in this paper and traditional methods such as installation of submerged vanes, and changing the inlet geometry (angle, radius) was performed. Figure 8 shows the comparison of the results. The comparison shows that the new techniques can be highly influential and still practical. In this research, with no change in structural geometry (enhancement of roughness coefficient) or minor changes with respect to drop implementation, the dimensions of the separation zone are decreased noticeably. The velocity values were also recorded by a one-dimensional velocity meter at 15 cm distance from the turnout entrance and in a transverse direction (perpendicular to the flow direction). The results are shown in Figure 9.Figure 9VIEW LARGEDOWNLOAD SLIDE

Effect of roughness on separation zone dimensions in numerical study.

Numerical results

Listen

This study examined the flow patterns around the entrance of a diversion channel due to various wall roughnesses in the diversion channel. Results indicated that increasing the discharge ratio in the main channel and diversion channel reduces the area of the separation zone in the diversion channel.Figure 10VIEW LARGEDOWNLOAD SLIDE

Comparision of the vortex area (software output) for three roughnesses (0.009, 0.023 and 0.032).A laboratory and numerical error rate of 0.2605 was calculated from the following formula,

formula

where Uexp is the experimental result, Unum is the numerical result, and N is the number of data.

Figure 9 shows the effect of roughness on separation zone dimensions in numerical study. Figure 10 compares the vortex area (software output) for three roughnesses, 0.009, 0.023 and 0.032 and Figure 11 shows the flow lines (tecplot output) that indicate the effect of roughness on flow in the separation zone. Numerical analysis shows that by increasing the roughness coefficient, the dimensions of the separation zone area decrease, as shown in Figure 10 where the separation zone area at n=0.032 is less than the separation zone area at n=0.009.Figure 11VIEW LARGEDOWNLOAD SLIDE

Comparison of vortex area in 3D mode (tecplot output) with two roughnesses (a) 0.009 and (b) 0.032.Figure 12VIEW LARGEDOWNLOAD SLIDE

Velocity vector for flow condition Q1/422 l/s, near surface.

The velocities intensified moving midway toward the turnout showing that the effective area is scaled down. The velocity values were almost equal to zero near the side walls as expected. As shown in Figure 12 the approach vortex area velocity decreases. Experimental and numerical measured velocity at x=0.15 m of the diversion channel compared in Figure 13 shows that away from the separation zone area, the velocity increases. All longitudinal velocity contours near the vortex area are distinctly different between different roughnesses. The separation zone is larger at less roughness both in length and width.Figure 13VIEW LARGEDOWNLOAD SLIDE

Exprimental and numerical measured velocity.

CONCLUSION

Listen

This study introduces practical and feasible methods for enhancing turnout efficiency by reducing the separation zone dimensions. Increasing the roughness coefficient and implementation of inlet drop were considered as remedies for reduction of separation zone dimensions. A data set has been compiled that fully describes the complex, 3D flow conditions present in a 90 degree turnout channel for selected flow conditions. The aim of this numerical model was to compare the results of a laboratory model in the area of the separation zone and velocity. Results showed that enhancing roughness coefficient reduce the separation zone dimensions up to 38% while the drop implementation effect can scale down this area differently based on roughness coefficient used. Combining both methods can reduce the separation zone dimensions up to 63%. Further research is proposed to investigate the effect of roughness and drop implementation on sedimentation pattern at lateral turnouts. The dimensions of the separation zone decreases with the increase of the non-dimensional parameter, due to the reduction ratio of turnout discharge increasing in all the experiments.

This method increases the discharge ratio (ratio of turnout to main channel discharge). The results are compatible with the literature. Other researchers have reported that intensifying the discharge ratio can scale down the separation zone dimensions (Karami Moghaddam & Keshavarzi 2007Ramamurthy et al. 2007). However, they employed other methods to enhance the discharge ratio. Employing both techniques simultaneously can decrease the separation zone dimensions up to 63%. A comparison between the new methods introduced in this paper and traditional methods such as installation of submerged vanes, and changing the inlet geometry (angle, radius) was performed. The comparison shows that the new techniques can be highly influential and still practical. The numerical and laboratory models are in good agreement and show that the method used in this study has been effective in reducing the separation area. This method is simple, economical and can prevent sediment deposition in the intake canal. Results show that CFD prediction of the fluid through the separation zone at the canal intake can be predicted reasonably well and the RNG model offers the best results in terms of predictability.

DATA AVAILABILITY STATEMENT

Listen

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

REFERENCES

Abbasi A., Ghodsian M., Habibi M. & Salehi Neishabouri S. A. 2004 Experimental investigation on dimensions of flow separation zone at lateral intakeentrance. Research & Construction; Pajouhesh va Sazandegi 62, 38–44. (In Persian).Google Scholar Al-Zubaidy R. & Hilo A. 2021 Numerical investigation of flow behavior at the lateral intake using Computational Fluid Dynamics (CFD). Materials Today: Proceedings. https://doi.org/10.1016/j.matpr.2021.11.172.Google Scholar Chow V. T. 1959 Open Channel Hydraulics. McGraw-Hill, New York.Jalili H., Hosseinzadeh Dalir A. & Farsadizadeh D. 2011 Effect of intake geometry on the sediment transport and lateral flow pattern. Iranian Water Research Journal 5 (9), 1–10. (In Persian).Google Scholar Jamshidi A., Farsadizadeh D. & Hosseinzadeh Dalir A. 2016 Variations of flow separation zone at lateral intake entrance using submerged vanes. Journal of Civil Engineering Urban 6 (3), 54–63. Journal homepage. Available from: www.ojceu.ir/main.Google Scholar Karami Moghaddam K. & Keshavarzi A. 2007 Investigation of flow structure in lateral intakes of 55° and 90° with rounded entrance edge. In: 03 National Congress on Civil Engineering University of Tabriz. Available from: https://civilica.com/doc/16317. (In Persian).Google Scholar Karami H., Farzin S., Sadrabadi M. T. & Moazeni H. 2017 Simulation of flow pattern at rectangular lateral intake with different dike and submerged vane scenarios. Journal of Water Science and Engineering 10 (3), 246–255. https://doi.org/10.1016/j.wse.2017.10.001.Google ScholarCrossref  Kasthuri B. & Pundarikanthan N. V. 1987 Discussion on separation zone at open- channel junction. Journal of Hydraulic Engineering 113 (4), 543–548.Google ScholarCrossref  Keshavarzi A. & Habibi L. 2005 Optimizing water intake angle by flow separation analysis. Journal of Irrigation and Drain 54, 543–552. https://doi.org/10.1002/ird.207.Google ScholarCrossref  Kirkgöz M. S. & Ardiçlioğlu M. 1997 Velocity profiles of developing and developed open channel flow. Journal of Hydraulic Engineering 1099–1105. 10.1061/(ASCE)0733-9429(1997)123:12(1099).Google Scholar Nakato T., Kennedy J. F. & Bauerly D. 1990 Pumpstation intake-shoaling control with submerge vanes. Journal of Hydraulic Engineering. https://doi.org/10.1061/(ASCE)0733-9429(1990)116:1(119).Google Scholar Neary V. S. & Odgaard J. A. 1993 Three-dimensional flow structure at open channel diversions. Journal of Hydraulic Engineering. ASCE 119 (11), 1224–1230. https://doi.org/10.1061/(ASCE)0733-9429(1993)119:11(1223).Google ScholarCrossref  Nikbin S. & Borghei S. M. 2011 Experimental investigation of submerged vanes effect on dimensions of flow separation zone at a 90° openchannel junction. In: 06rd National Congress on Civil Engineering University of Semnan. (In Persian). Available from: https://civilica.com/doc/120494.Google Scholar Odgaard J. A. & Wang Y. 1991 Sediment management with submerged vanes, I: theory. Journal of Hydraulic Engineering 117 (3), 267–283.Google ScholarCrossref  Ramamurthy A. S., Junying Q. & Diep V. 2007 Numerical and experimental study of dividing open-channel flows. Journal of Hydraulic Engineering. See: https://doi.org/10.1061/(ASCE)0733-9429(2007)133:10(1135).Google Scholar Seyedian S., Karami Moghaddam K. & Shafai Begestan M. 2008 Determining the optimal radius in lateral intakes of 55° and 90° using variation of flow velocity. In: 07th Iranian Hydraulic Conference. Power & Water University of Technology (PWUT). (In Persian). Available from: https://civilica.com/doc/56251.Google Scholar Zolghadr M. & Shafai Bejestan M. 2020 Six legged concrete (SLC) elements as scour countermeasures at wing wall bridge abutments. International Journal of River Basin Management. doi: 10.1080/15715124.2020.1726357.Google Scholar Zolghadr M., Zomorodian S. M. A., Shabani R. & Azamatulla H.Md. 2021 Migration of sand mining pit in rivers: an experimental, numerical and case study. Measurement. https://doi.org/10.1016/j.measurement.2020.108944.Google Scholar © 2022 The AuthorsThis is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition

Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition

Xiang WangLin-Jie ZhangJie Ning, and Suck-Joo Na
Published Online:8 Apr 2022https://doi.org/10.1089/3dp.2021.0159

Abstract

A 3D numerical model of heat transfer and fluid flow of molten pool in the process of laser wire deposition was presented by computational fluid dynamics technique. The simulation results of the deposition morphology were also compared with the experimental results under the condition of liquid bridge transfer mode. Moreover, they showed a good agreement. Considering the effect of recoil pressure, the morphology of the deposit metal obtained by the simulation was similar to the experiment result. Molten metal at the wire tip was peeled off and flowed into the molten pool, and then spread to both sides of the deposition layer under the recoil pressure. In addition, the results of simulation and high-speed charge-coupled device presented that a wedge transition zone, with a length of ∼6 mm, was formed behind the keyhole in the liquid bridge transfer process, where the height of deposited metal decreased gradually. After solidification, metal in the transition zone retained the original melt morphology, resulting in a decrease in the height of the tail of the deposition layer.

Keywords

LWD, CFD, liquid bridge transfer, fluid dynamics, wedge transition zone

Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition
Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition
Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition
Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition

References

1. Matthews MJ, Guss G, Khairallah SA, et al. Denudation of metal powder layers in laser powder bed fusion processes. Acta Mater 2016;114:33–42. CrossrefGoogle Scholar

2. Ge WJ, Han SW, Fang YC, et al. Mechanism of surface morphology in electron beam melting of Ti6Al4V based on computational flow patterns. Appl Surf Sci 2017;419:150–158. CrossrefGoogle Scholar

3. Bai XW, Colegrove P, Ding JL, et al. Numerical analyswas of heat transfer and fluid flow in multilayer deposition of PAW-based wire and arc additive manufacturing. Int J Heat Mass Transf 2018;124:504–516. CrossrefGoogle Scholar

4. Torkamany MJ, Kaplan AFH, Ghaini FM. Wire deposition by a laser-induced boiling front. Opt Laser Technol 2015;69:104–112. CrossrefGoogle Scholar

5. Yu Y, Huang W, Wang G. Investigation of melting dynamics of filler wire during wire feed laser welding. J Mec Sci Technol 2013;27:1097–1108. CrossrefGoogle Scholar

6. Ma G, Li L, Chen Y. Effects of beam confgurations on wire melting and transfer behaviors in dual beam laser welding with fller wire. Opt Laser Technol 2017;91:138–148. CrossrefGoogle Scholar

7. Abioye TE, Folkes J, Clare AT. A parametric study of Inconel 625 wire laser deposition. J Mater Process Tech 2013;213:2145–2151. CrossrefGoogle Scholar

8. Wei S, Wang G, Shin YC, et al. Comprehensive modeling of transport phenomena in laser hot-wire deposition process. Int J Heat Mass Transf 2018;125:1356–1368. CrossrefGoogle Scholar

9. Gu H, Li L. Computational fluid dynamic simulation of gravity and pressure effects in laser metal deposition for potential additive manufacturing in space. Int J Heat Mass Transf 2019;140:51–65. CrossrefGoogle Scholar

10. Hu R, Luo M, Liu T, et al. Thermal fluid dynamics of liquid bridge transfer in laser wire deposition 3D printing. Sci Technolf Weld Join 2019;24:1–11. Google Scholar

11. Chatterjee D, Chakraborty S. A hybrid lattice Boltzmann model for solid–liquid phase transition in presence of fluid flow. Phys Lett A 2006;351:359–367. CrossrefGoogle Scholar

12. Wu L, Cheon J, Kiran DV, et al. CFD simulations of GMA welding of horizontal fillet joints based on coordinate rotation of arc models. J Mater Process Tech 2016;231:221–238. CrossrefGoogle Scholar

13. Gerhard W, Boyer RR, Collings EW. Materials Properties Handbook: Titanium Alloys. ASM International: Almere, The Netherlands, 1994. Google Scholar

14. Colegrove P, Simiand PE, Varughese A, et al. Evaluation of a drilling model approach to represent laser spot microwelding. In: ASM Proceedings of the international conference: trends in welding research; 2009. Google Scholar

15. Boivineau M, Cagran C, Doytier D, et al. Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy. Int J Thermophys 2006;27:507–529. CrossrefGoogle Scholar

16. Shejndlin AE, Kenisarin MM, Chekhovskoj VY. Melting point of yttrium oxide. AN SSSR 1974;216:582–584. Google Scholar

17. Cho JH, Na SJ. Teflection and Fresnel absorption of laser beam in keyhole. J Phys D Appl Phys 2006;39:5372–5378. CrossrefGoogle Scholar

18. Han SW, Ahn J, Na SJ. A study on ray tracing method for CFD simulations of laser keyhole welding: Progressive search method. Weld World 2016;60:247–258. CrossrefGoogle Scholar

19. Allmen MV. Laser-Beam Interactions with Materials. Springer, Berlin-Heidelberg, 1995. Google Scholar

20. Dobson PJ. Absorption and scattering of light by small particles. Phys Bull 1984;35:104. CrossrefGoogle Scholar

21. Greses J, Hilton PA, Barlow CY. Plume attenuation under high power Nd:yttritium aluminum garnet laser welding. J Laser Appl 2004;16:9–15. CrossrefGoogle Scholar

22. Shcheglov PY, Uspenskiy SA, Gumenyuk AV, et al. Plume attenuation of laser radiation during high power fiber laser welding. Laser Phys Lett 2011;8:475–480. CrossrefGoogle Scholar

23. Yang P, Liou KN. Effective refractive index for determining ray propagation in an absorbing dielectric particle. J Quant Spectrosc Radiat Transf 2009;110:300–306. CrossrefGoogle Scholar

24. Barber PW. Absorption and scattering of light by small particles. J Colloid Interface Sci 1984;98:290–291. Google Scholar

25. Hu ZR, Chen X, Yang G, et al. Metal transfer in wire feeding-based electron beam 3D printing: Modes, dynamics, and transition criterion. Int J Heat Mass Transf 2018;126:877–887. CrossrefGoogle Scholar

26. David SA, Babu SS, Vitek JM. Welding: Solidification and microstructure. JOM 2013;55:14–20. CrossrefGoogle Scholar

27. Zhong ML, Liu W. Laser surface cladding: The state of the art and challenges. Proc Inst Mech Eng Part C J Mech Eng Sci 2010;224:1041–1060. CrossrefGoogle Scholar

28. Kobryn PA, Semiatin S. Microstructure and texture evolution during solidification processing of Ti-6Al-4V. J Mater Process Technol 2003;135:330–339. CrossrefGoogle Scholar

29. Debroy T, David S. Physical processes in fusion welding. Rev Mod Phys 1995;67:85–112. CrossrefGoogle Scholar

30. Lee YS, Nordin M, Babu SS, et al. Effect of fluid convection on dendrite arm spacing in laser deposition. Metall Trans B 2014;45:1520–1528. CrossrefGoogle Scholar

31. Rappaz M, David SA, Vitek JM, et al. Development of microstructures in Fe15Ni15Cr single crystal electron beam welds. Metall Trans A 1989;20:1125–1138. CrossrefGoogle Scholar

Forming characteristics and control method of weld bead for GMAW on curved surface

곡면에 GMAW용 용접 비드의 형성 특성 및 제어 방법

Forming characteristics and control method of weld bead for GMAW on curved surface

The International Journal of Advanced Manufacturing Technology (2021)Cite this article

Abstract

곡면에서 GMAW 기반 적층 가공의 용접 성형 특성은 중력의 영향을 크게 받습니다. 성형면의 경사각이 크면 혹 비드(hump bead)와 같은 심각한 결함이 발생합니다.

본 논문에서는 양생면에서 용접 비드 형성의 형성 특성과 제어 방법을 연구하기 위해 용접 용융 풀 유동 역학의 전산 모델을 수립하고 제안된 모델을 검증하기 위해 증착 실험을 수행하였습니다.

결과는 용접 비드 경사각(α)이 증가함에 따라 역류의 속도가 증가하고 상향 용접의 경우 α > 60°일 때 불규칙한 험프 결함이 나타나는 것으로 나타났습니다.

상부 과잉 액체의 하향 압착력과 하부 상향 유동의 반동력과 표면장력 사이의 상호작용은 용접 혹 형성의 주요 요인이었다. 하향 용접의 경우 양호한 형태를 얻을 수 있었으며, 용접 비드 경사각이 증가함에 따라 용접 높이는 감소하고 용접 폭은 증가하였습니다.

하향 및 상향 용접을 위한 곡면의 용융 거동 및 성형 특성을 기반으로 험프 결함을 제어하기 위해 위브 용접을 통한 증착 방법을 제안하였습니다.

성형 궤적의 변화로 인해 용접 방향의 중력 성분이 크게 감소하여 용융 풀 흐름의 안정성이 향상되었으며 복잡한 표면에서 안정적이고 일관된 용접 비드를 얻는 데 유리했습니다.

하향 용접과 상향 용접 사이의 단일 비드의 치수 편차는 7% 이내였으며 하향 및 상향 혼합 혼합 비드 중첩 증착에서 비드의 변동 편차는 0.45로 GMAW 기반 적층 제조 공정에서 허용될 수 있었습니다.

이러한 발견은 GMAW를 기반으로 하는 곡선 적층 적층 제조의 용접 비드 형성 제어에 기여했습니다.

The weld forming characteristics of GMAW-based additive manufacturing on curved surface are dramatically influenced by gravity. Large inclined angle of the forming surface would lead to severe defects such as hump bead. In this paper, a computational model of welding molten pool flow dynamics was established to research the forming characteristic and control method of weld bead forming on cured surface, and deposition experiments were conducted to verify the proposed model. Results indicated that the velocity of backward flows increased with the increase of weld bead tilt angle (α) and irregular hump defects appeared when α > 60° for upward welding. The interaction between the downward squeezing force of the excess liquid at the top and the recoil force of the upward flow at the bottom and the surface tension were primary factors for welding hump formation. For downward welding, a good morphology shape could be obtained, and the weld height decreased and the weld width increased with the increase of weld bead tilt angle. Based on the molten behaviors and forming characteristics on curved surface for downward and upward welding, the method of deposition with weave welding was proposed to control hump defects. Gravity component in the welding direction was significantly reduced due to the change of forming trajectory, which improved the stability of the molten pool flow and was beneficial to obtain stable and consistent weld bead on complex surface. The dimensional deviations of the single bead between downward and upward welding were within 7% and the fluctuation deviation of the bead in multi-bead overlapping deposition with mixing downward and upward welding was 0.45, which could be acceptable in GMAW-based additive manufacturing process. These findings contributed to the weld bead forming control of curve layered additive manufacturing based on GMAW.

Keywords

  • Molten pool behaviors
  • GMAW-based WAAM
  • Deposition with weave welding
  • Welding on curved surface
  • Fig. 1extended data figure 1
  • Fig. 2extended data figure 2
  • Fig. 3extended data figure 3
  • Fig. 4extended data figure 4
  • Fig. 5extended data figure 5
  • Fig. 6extended data figure 6
  • Fig. 7extended data figure 7
  • Fig. 8extended data figure 8
  • Fig. 9extended data figure 9
  • Fig. 10extended data figure 10
  • Fig. 11extended data figure 11
  • Fig. 12extended data figure 12
  • Fig. 13extended data figure 13
  • Fig. 14extended data figure 14
  • Fig. 15extended data figure 15
  • Fig. 16extended data figure 16
  • Fig. 17extended data figure 17
  • Fig. 18extended data figure 18
  • Fig. 19extended data figure 19
  • Fig. 20extended data figure 20
  • Fig. 21extended data figure 21
  • Fig. 22extended data figure 22
  • Fig. 23extended data figure 23
  • Fig. 24extended data figure 24
  • Fig. 25extended data figure 25
  • Fig. 26extended data figure 26
  • Fig. 27extended data figure 27
  • Fig. 28extended data figure 28
  • Fig. 29extended data figure 29
  • Fig. 30extended data figure 30
  • Fig. 31extended data figure 31
  • Fig. 32extended data figure 32
  • Fig. 33extended data figure 33
  • Fig. 34extended data figure 34
  • Fig. 35extended data figure 35
  • Fig. 36extended data figure 36
  • Fig. 37extended data figure 37
  • Fig. 38extended data figure 38

References

  1. 1.Williams SW, Martina F, Addison AC, Ding J, Pardal G, Colegrove P (2016) Wire + arc additive manufacturing. Mater Sci Technol (United Kingdom) 32:641–647. https://doi.org/10.1179/1743284715Y.0000000073Article Google Scholar 
  2. 2.Pan ZX, Ding DH, Wu BT, Cuiuri D, Li HJ, Norrish J (2018) Arc welding processes for additive manufacturing: a review. In: Transactions on intelligent welding manufacturing. Springer Singapore, pp 3–24. https://doi.org/10.1007/978-981-10-5355-9_1
  3. 3.Panchagnula JS, Simhambhatla S (2018) Manufacture of complex thin-walled metallic objects using weld-deposition based additive manufacturing. Robot Comput Integr Manuf 49:194–203. https://doi.org/10.1016/j.rcim.2017.06.003Article Google Scholar 
  4. 4.Lu S, Zhou J, Zhang JS (2015) Optimization of welding thickness on casting-steel surface for production of forging die. Int J Adv Manuf Technol 76:1411–1419. https://doi.org/10.1007/s00170-014-6371-9Article Google Scholar 
  5. 5.Huang B, Singamneni SB (2015) Curved layer adaptive slicing (CLAS) for fused deposition modelling. Rapid Prototyp J 21:354–367. https://doi.org/10.1108/RPJ-06-2013-0059Article Google Scholar 
  6. 6.Jin Y, Du J, He Y, Fu GQ (2017) Modeling and process planning for curved layer fused deposition. Int J Adv Manuf Technol 91:273–285. https://doi.org/10.1007/s00170-016-9743-5Article Google Scholar 
  7. 7.Xie FB, Chen LF, Li ZY, Tang K (2020) Path smoothing and feed rate planning for robotic curved layer additive manufacturing. Robot Comput Integr Manuf 65. https://doi.org/10.1016/j.rcim.2020.101967
  8. 8.Ding YY, Dwivedi R, Kovacevic R (2017) Process planning for 8-axis robotized laser-based direct metal deposition system: a case on building revolved part. Robot Comput Integr Manuf 44:67–76. https://doi.org/10.1016/j.rcim.2016.08.008Article Google Scholar 
  9. 9.Cho DW, Na SJ (2015) Molten pool behaviors for second pass V-groove GMAW. Int J Heat Mass Transf 88:945–956. https://doi.org/10.1016/j.ijheatmasstransfer.2015.05.021Article Google Scholar 
  10. 10.Cho DW, Na SJ, Cho MH, Lee JS (2013) A study on V-groove GMAW for various welding positions. J Mater Process Technol 213:1640–1652. https://doi.org/10.1016/j.jmatprotec.2013.02.015Article Google Scholar 
  11. 11.Hejripour F, Valentine DT, Aidun DK (2018) Study of mass transport in cold wire deposition for wire arc additive manufacturing. Int J Heat Mass Transf 125:471–484. https://doi.org/10.1016/j.ijheatmasstransfer.2018.04.092Article Google Scholar 
  12. 12.Yuan L, Pan ZX, Ding DH, He FY, Duin SV, Li HJ, Li WH (2020) Investigation of humping phenomenon for the multi-directional robotic wire and arc additive manufacturing. Robot Comput Integr Manuf 63. https://doi.org/10.1016/j.rcim.2019.101916
  13. 13.Nguyen MC, Medale M, Asserin O, Gounand S, Gilles P (2017) Sensitivity to welding positions and parameters in GTA welding with a 3D multiphysics numerical model. Numer Heat Transf Part A Appl 71:233–249. https://doi.org/10.1080/10407782.2016.1264747Article Google Scholar 
  14. 14.Gu H, Li L (2019) Computational fluid dynamic simulation of gravity and pressure effects in laser metal deposition for potential additive manufacturing in space. Int J Heat Mass Transf 140:51–65. https://doi.org/10.1016/j.ijheatmasstransfer.2019.05.081Article Google Scholar 
  15. 15.Cho MH, Farson DF (2007) Understanding bead hump formation in gas metal arc welding using a numerical simulation. Metall Mater Trans B Process Metall Mater Process Sci 38:305–319. https://doi.org/10.1007/s11663-007-9034-5Article Google Scholar 
  16. 16.Nguyen TC, Weckman DC, Johnson DA, Kerr HW (2005) The humping phenomenon during high speed gas metal arc welding. Sci Technol Weld Join 10:447–459. https://doi.org/10.1179/174329305X44134Article Google Scholar 
  17. 17.Philip Y, Xu ZY, Wang Y, Wang R, Ye X (2019) Investigation of humping defect formation in a lap joint at a high-speed hybrid laser-GMA welding. Results Phys 13. https://doi.org/10.1016/j.rinp.2019.102341
  18. 18.Hu ZQ, Qin XP, Shao T, Liu HM (2018) Understanding and overcoming of abnormity at start and end of the weld bead in additive manufacturing with GMAW. Int J Adv Manuf Technol 95:2357–2368. https://doi.org/10.1007/s00170-017-1392-9Article Google Scholar 
  19. 19.Tang SY, Wang GL, Huang C, Li RS, Zhou SY, Zhang HO (2020) Investigation, modeling and optimization of abnormal areas of weld beads in wire and arc additive manufacturing. Rapid Prototyp J 26:1183–1195. https://doi.org/10.1108/RPJ-08-2019-0229Article Google Scholar 
  20. 20.Bai X, Colegrove P, Ding J, Zhou XM, Diao CL, Bridgeman P, Honnige JR, Zhang HO, Williams S (2018) Numerical analysis of heat transfer and fluid flow in multilayer deposition of PAW-based wire and arc additive manufacturing. Int J Heat Mass Transf 124:504–516. https://doi.org/10.1016/j.ijheatmasstransfer.2018.03.085Article Google Scholar 
  21. 21.Siewert E, Schein J, Forster G (2013) Determination of enthalpy, temperature, surface tension and geometry of the material transfer in PGMAW for the system argon-iron. J Phys D Appl Phys 46. https://doi.org/10.1088/0022-3727/46/22/224008
  22. 22.Goldak J, Chakravarti A, Bibby M (1984) A new finite element model for welding heat sources. Metall Trans B 15:299–305. https://doi.org/10.1007/BF02667333Article Google Scholar 
  23. 23.Fachinotti VD, Cardona A (2008) Semi-analytical solution of the thermal field induced by a moving double-ellipsoidal welding heat source in a semi-infinite body. Mec Comput XXVII:1519–1530
  24. 24.Nguyen NT, Mai YW, Simpson S, Ohta A (2004) Analytical approximate solution for double ellipsoidal heat source in finite thick plate. Weld J 83:82–93Google Scholar 
  25. 25.Goldak J, Chakravarti A, Bibby M (1985) A double ellipsoid finite element model for welding heat sources. IIW Doc. No. 212-603-85
  26. 26.Gu Y, Li YD, Yong Y, Xu FL, Su LF (2019) Determination of parameters of double-ellipsoidal heat source model based on optimization method. Weld World 63:365–376. https://doi.org/10.1007/s40194-018-00678-wArticle Google Scholar 
  27. 27.Wu CS, Tsao KC (1990) Modelling the three-dimensional fluid flow and heat transfer in a moving weld pool. Eng Comput 7:241–248. https://doi.org/10.1108/eb023811Article Google Scholar 
  28. 28.Zhan XH, Liu XB, Wei YH, Chen JC, Chen J, Liu HB (2017) Microstructure and property characteristics of thick Invar alloy plate joints using weave bead welding. J Mater Process Technol 244:97–105. https://doi.org/10.1016/j.jmatprotec.2017.01.014Article Google Scholar 
  29. 29.Zhan XH, Zhang D, Liu XB, Chen J, Wei YH, Liu RP (2017) Comparison between weave bead welding and multi-layer multi-pass welding for thick plate Invar steel. Int J Adv Manuf Technol 88:2211–2225. https://doi.org/10.1007/s00170-016-8926-4Article Google Scholar 
  30. 30.Xu GX, Li L, Wang JY, Zhu J, Li PF (2018) Study of weld formation in swing arc narrow gap vertical GMA welding by numerical modeling and experiment. Int J Adv Manuf Technol 96:1905–1917. https://doi.org/10.1007/s00170-018-1729-zArticle Google Scholar 
  31. 31.Li YZ, Sun YF, Han QL, Zhang GJ, Horvath I (2018) Enhanced beads overlapping model for wire and arc additive manufacturing of multi-layer multi-bead metallic parts. J Mater Process Technol 252:838–848. https://doi.org/10.1016/j.jmatprotec.2017.10.017Article Google Scholar 
Heat and Mass Transfer in a Cryogenic Tank in Case of Active-Pressurization

능동 가압의 경우 극저온 탱크의 열 및 물질 전달

Heat and Mass Transfer in a Cryogenic Tank in Case of Active-Pressurization

하이라이트

헤닝 슈플러 옌스 게르스트만DLR 독일 항공 우주 센터, 우주 시스템 연구소, 28359 Bremen, Germany

상변화 및 공액 열전달을 포함하는 압축성 2상 솔버 개발.

분석 솔루션으로 솔버를 성공적으로 검증.

극저온 탱크의 압력 및 온도 변화에 대한 정확한 시뮬레이션.

자유 표면에서의 물질 전달 분석.

Abstract

압력 요구 사항을 예측하는 것은 극저온 추진 시스템의 주요 과제 중 하나입니다. 이러한 맥락에서 증발 및 응축 현상을 고려한 탱크 여압을 시뮬레이션하기 위한 수치 모델을 개발하여 적용하였습니다. 

새로운 솔버는 PISO(splitting of operator) 알고리즘이 있는 압력 암시적 방법을 기반으로 하는 OpenFOAM의 약한 압축성 다상 솔버와 기울기 기반 위상 변화 모델을 결합합니다. 날카로운 인터페이스를 유지하기 위해 인터페이스에 인접한 셀에 질량 소스 용어가 적용됩니다. 

첫째, 모델은 1차원 상 변화 문제와 중력이 없는 상태에서 과열된 액체에서 증기 기포의 성장이라는 두 가지 분석 솔루션에 대해 검증되었습니다. 

두 번째 단계에서는 검증된 모델을 극저온 가압 실험에 적용했습니다. 측정된 압력 거동은 수치 모델이 양호한 근사값으로 확인될 수 있습니다. 

수치 모델을 사용하면 물리적 거동에 대한 추가 통찰력을 얻을 수 있습니다. 응축 및 증발 효과는 가압 중 및 가압 후의 압력 발생에 상당한 영향을 미칩니다. 기액 계면에서 일어나는 상변화로 인한 질량유동은 계면의 위치와 시간에 따라 달라진다. 벽에서 직접적으로 증발이 지배적이며 액체 표면의 중앙 영역에서 응결이 발생합니다. 

응축 및 증발 효과는 가압 중 및 가압 후의 압력 발생에 상당한 영향을 미칩니다. 기액 계면에서 일어나는 상변화로 인한 질량유동은 계면의 위치와 시간에 따라 달라진다. 벽에서 직접적으로 증발이 지배적이며 액체 표면의 중앙 영역에서 응결이 발생합니다. 

응축 및 증발 효과는 가압 중 및 가압 후의 압력 발생에 상당한 영향을 미칩니다. 기액 계면에서 일어나는 상변화로 인한 질량유동은 계면의 위치와 시간에 따라 달라진다. 벽에서 직접적으로 증발이 지배적이며 액체 표면의 중앙 영역에서 응결이 발생합니다.

Predicting the pressurant requirements is one of the key challenges for cryogenic propulsion systems. In this context, a numerical model to simulate the tank pressurization that considers evaporation and condensation phenomena was developed and applied. The novel solver combines the a gradient-based phase change model with a weakly compressible multiphase solver of OpenFOAM based on the pressure implicit method with splitting of operator (PISO) algorithm. To maintain a sharp interface the mass source terms are applied to the cells adjacent to the interface. First, the model is validated against two analytical solutions: the one-dimensional phase change problem and secondly, the growth of a vapor bubble in a superheated liquid in the absence of gravity. In a second step, the validated model was applied to a cryogenic pressurization experiment. The measured pressure behavior could be confirmed with the numerical model being in a good approximation. With the numerical model further insights into the physical behavior could be achieved. The condensation and evaporation effects have a significant impact on the pressure development during and after the pressurization. The mass flows due to phase change occurring at the vapor-liquid interface depend on interface location and time. Directly at the wall, evaporation becomes dominant while condensation occurs at the center area of the liquid surface.

  1. Fig. 1. Calculation of the gradient at the interface: On the left side the interface…
  2. Fig. 2. Mass source term distribution: First the sharp mass source term ρ0, which is…
  3. Fig. 3. a) Layout of the Stefan-Problem: a vapor is located between a liquid and a…
  4. Fig. 4. Bubble in a superheated liquid: The left side depicts the calculated and…
  5. Fig. 5. Modified drawing of the dewar (as documented in [5] [6]; dimensions in mm) and…
  6. Fig. 6. Schematic presentation of the pressure evoluation in the dewar: Initial…
  7. Fig. 7. Simulation of the pressurization phase: The diagram shows the pressure…
  8. Fig. 8. Turbulent thermal diffusivity in pressurization and relaxation phase
  9. Fig. 9. Comparison of the pressure evolution in the relaxation phase of the solver with…
  10. Fig. 10. On the left side the temperature evolution in the bulk of the gas phase is shown
  11. Fig. 11. Heat Flux profile over the interface caused by evaporation with details of the…
  12. Fig. 12. Temperatures field with velocity vectors at 420 seconds after the start of the…
  13. Fig. 13. Heat transfer to the liquid from the wall and the freesurface with and without…

Hide figures

키워드

Pressurization, Phase Change, CFD, Propellant Management, 가압, 상 변화, 추진제 관리

그림 3. 수중 4차 횡파 영향

Validation of Sloshing Simulations in Narrow Tanks

This case study was contributed by Peter Arnold, Minerva Dynamics.

이 작업의 목적은 FLOW-3D  를 검증하는 것입니다. 밀폐된 좁은 스팬 직사각형 탱크의 출렁거림 문제에 대비하여 탱크의 내부 파동 공명 주기에 가깝거나 같은 주기로 롤 운동을 하여 측면 및 지붕 파동 충격 이벤트가 발생합니다.

탱크는 물이나 해바라기 기름으로 두 가지 다른 수준으로 채워졌고 위의 공간은 공기로 채워졌습니다. 압력 센서는 여러 장소의 벽에 설치되었으며 처음 4개의 출렁이는 기간 동안 기록된 롤 각도와 시간 이력이 있습니다. 오일을 사용하는 경우의 흐름은 레이놀즈 수가 1748인 층류인 반면, 물로 채워진 경우의 흐름은 레이놀즈 수가 97546인 난류입니다. 

CFD 시뮬레이션은 탱크의 고조파 롤 운동을 복제하기 위해 본체력 방법을 사용했으며, 난류 및 공기 압축성을 설명하기 위해 다른 모델링 가정과 함께 그리드 의존성 테스트를 수행했습니다.

The objective of this work is to validate FLOW-3D against a sloshing problem in a sealed narrow span rectangular tank, subjected to roll motion at periods close to or equal to the tank’s internal wave resonance period, such that side and roof wave impact events occur. The tank was filled to two different levels with water or sunflower oil, with the space above filled by air. Pressure sensors were installed in the walls at several places and their time histories, along with the roll angle, recorded for the first four sloshing periods. For the cases using oil, the flow is laminar with a Reynolds number of 1748, while for the cases filled with water the flow is turbulent with a Reynolds number of 97546. The CFD simulations used the body force method to replicate the harmonic roll motion of the tank, while grid dependence tests were performed along with different modelling assumptions to account for turbulence and air compressibility.

Experimental Problem Setup

원래 실험은 Souto-Iglesias 및 Botia-Vera[1]에 의해 수행되었으며 모든 실험 데이터 파일은 문제 설명, 비디오 및 불확실성 분석과 함께 사용할 수 있습니다. 그림 1에 표시된 형상은 길이 900mm, 높이 508mm, 스팬 62mm의 직사각형 탱크로 구성되어 있으며 물이나 해바라기 기름으로 93mm 또는 355.3mm로 채워져 있으므로 4가지 경우가 고려됩니다. 탱크 벽과 같은 높이로 설치된 압력 센서의 위치도 표시됩니다. 탱크 회전 중심은 수평에 대한 회전 각도와 함께 그림 1에 나와 있습니다. 각 실험 실행은 반복성을 평가할 수 있도록 100번 수행되었습니다.

The original experiment was performed by Souto-Iglesias and Botia-Vera [1] and all experimental data files are available along with problem description, videos and an uncertainty analysis. The geometry shown in Fig. 1 consists of a rectangular tank of 900mm length, 508mm height and 62mm span, filled to either 93mm or 355.3 mm with either water or sunflower oil, hence four cases are considered. The locations of the pressure sensors that were installed flush with the tank walls are also shown. The tank rotation center is shown in Fig. 1, along with the rotation angle relative to the horizontal. Each of the experimental runs was performed 100 times to enable their repeatability to be assessed.

Tank dimensions and locations of pressure sensors
Figure 1. Tank dimensions and locations of pressure sensors

Numerical Simulation

문제는 FLOW-3D 내에서 비관성 기준 좌표계 모델을 사용하여 비교적 간단하게 설정할 수 있으며  , 이는 로컬 기준 좌표계의 가속도에 따라 유체에 체력 을 적용합니다. Z축 회전 속도는 탱크의 롤 운동을 시뮬레이션하기 위한 주기 함수로 정의되었으며 음의 수직 방향으로 작용하는 일정한 중력이 가해졌습니다.

메쉬 미세화, 운동량 이류에 대한 수치 근사 순서, 층류 대 난류 모델 및 탱크 내 공기에 대한 세 가지 다른 처리(즉, 일정 압력, 압축성 기체 및 비압축성 기체)와 같은 것을 조사하기 위해 여러 시뮬레이션을 수행했습니다.

93mm 깊이로 채워진 모든 케이스에 대해 압력은 압력 센서 P1에서만 실험 값과 비교되었으며, 355.3mm 깊이로 채워진 모든 케이스에서는 P3 센서의 데이터만 비교되었습니다.

The problem was relatively simple to set up using the non-inertial reference frame model within FLOW-3D, which applies a body force to the fluid depending on the acceleration of the local reference frame. The Z axis rotational velocity was defined as a periodic function to simulate a roll motion of the tank, and a constant gravity force acting in the negative vertical direction was applied.

Multiple simulations were performed to investigate such things as mesh refinement, the numerical approximation order for momentum advection, laminar versus turbulent models and three different treatments for the air in the tank (i.e., constant pressure, compressible gas and incompressible gas).

For all 93mm depth-filled cases, the pressure was compared to the experimental values at pressure sensor P1 only, while for all 355.3mm depth-filled cases, only data at the P3 sensor was compared.

Results

P1에서 측정된 측면 워터 슬로싱에 대한 메쉬 해상도의 영향은 그림 2에서 볼 수 있습니다. 피크 값 예측 측면에서 특별한 편향을 보이지 않습니다. 모든 측면 사례에서 초기 피크 직후의 압력은 시뮬레이션에서 일관되게 과대 평가되었습니다. 모든 메쉬는 피크의 타이밍 측면에서 우수한 일치를 보입니다. 100회 실행에서 보고된 실험 시간 기록은 평균 값에 가장 가까운 최고 압력을 가진 기록입니다.

The effect of mesh resolution on lateral water sloshing measured at P1 is seen in Fig. 2. It shows no particular bias in terms of the prediction of peak values. In all the Lateral cases, the pressures immediately after the initial peaks are consistently over estimated in the simulations. All meshes have excellent agreement in terms of the timing of the peaks. The experimental time histories reported from the 100 runs made are those with peak pressures closest to the average values.

Lateral water case
Figure 2. Tank dimensions and locations of pressure sensors

실험 결과의 반복성은 Souto-Iglesias & Elkin Botia-Vera[1]에 의해 각 테스트를 100번 실행하고 처음 4개의 피크 압력의 평균 및 표준 편차를 측정하여 평가했습니다. CFD 실행이 다른 실험 실행으로 간주되는 경우 오류 막대 내에 있을 확률이 95%입니다. 그러나 CFD 결과의 16개 피크 압력 중 9개만 실험 결과의 2 표준 편차 내에 있으므로 CFD 모델이 실험을 대표하지 않거나 피크 압력이 정규 분포를 따르지 않는다는 결론을 내려야 합니다.

어쨌든 표준 편차는 피크 자체에 비해 상당히 크며, 수성 케이스와 측면 오일의 비율이 가장 작은 피크 값에 대한 표준 편차의 비율이 가장 큰 것으로 나타났습니다. 이러한 결과는 그림 1과 2에서 볼 수 있는 벽 충격 역학의 복잡성을 고려할 때 그리 놀라운 일이 아닙니다. 3,4.

The repeatability of the experimental results was assessed by Souto-Iglesias & Elkin Botia-Vera [1] running each test 100 times and measuring the average and standard deviation of the first four peak pressures. If a CFD run is considered to be another experimental run there is a 95% chance it will lie within the error bars. However, only nine of the 16 peak pressures from the CFD results fall within two standard deviations of the experimental results, so we must conclude that either the CFD model is not representative of the experiment or that the peak pressures are not normally distributed.

In any event, the standard deviations are quite large compared to the peaks themselves, with the largest ratio of standard deviation to peak values occurring for the water-based cases and the lateral oil having the smallest ratio. These results are perhaps not too surprising when one considers the complexity of the wall impact dynamics as seen in Figs. 3,4.

Lateral Wave Impact in Water
Figure 3. 4th Lateral Wave Impact in Water
Wave Impact of Water on Roof
Figure 4. 4th Wave Impact of Water on Roof

Conclusions

좁은 탱크 슬로싱 문제의 네 가지 구성은 자유 표면 흐름을 위해 설계된 상용 CFD 코드를 사용하여 수치적으로 시뮬레이션되었습니다. 대략 2 X 10 3  및 1 X 10 5 의 Reynolds 수에 해당하는 두 가지 다른 유체  와 두 가지 유체 깊이가 네 가지 경우를 정의하는 데 사용되었습니다. 4가지 경우 모두에 대해 메쉬 셀 크기 독립성 테스트를 수행했지만 메쉬 해상도가 증가함에 따라 실험 결과에 대해 약한 수렴만 발견되었습니다. 조사는 또한 두 가지 다른 운동량 이류 수치 차분 계획을 테스트했으며 두 번째 방법을 사용하여 더 가까운 일치를 발견했습니다 1차 체계를 사용하는 것보다 차수 단조성 보존 체계. 기본 층류 흐름을 포함한 세 가지 난류 모델이 테스트되었지만 더 낮은 계산 비용으로 인해 층류 이외의 모델에 대한 선호도가 발견되지 않았습니다. 실험 데이터와 공기 감소 일치의 압축성을 포함하여 그 이유는 불분명합니다.

실험 압력 프로브 시간 이력 데이터 세트에는 100회 반복 테스트에서 파생된 각 압력 피크에 대해 100개의 값이 포함되어 있으므로 CFD 시뮬레이션과의 일치의 통계적 유의성을 조사할 수 있었습니다. 수치 시뮬레이션과 실험 모두 출렁이는 파동 충격에 해당하는 매우 가파른 압력 펄스를 발생시켰고 실험 결과는 피크 값에서 높은 정도의 자연적 변동성을 갖는 것으로 나타났습니다. CFD 시뮬레이션의 감도 테스트(예: 약간 다른 초기 시작 조건 사용)는 공식적으로 수행되지 않았지만 수치 솔루션은 또한 다른 메쉬, 차분 체계 및 난류 모델,

모든 경우에 압력 피크가 발생하는 수치해의 타이밍은 매우 정확함을 알 수 있었다. 그러나 가장 난이도가 낮은 Lateral Oil의 경우에도 압력 피크와 바로 뒤따르는 압력 값이 과대 평가되어 수치 모델링의 단점이 나타났습니다. 실험적 피크 압력 변동성을 고려할 때 CFD 생성 값은 CFD 솔루션이 통계적 유의성을 나타내기 위해 필요한 15개 이상이 아니라 16개 피크 중 9개에서 2개의 표준편차 한계 내에 떨어졌습니다. 실험을 대표했다. 이것은 피크가 정규 분포를 따르지 않거나 CFD 모델이 피크를 예측하는 데 어떤 식으로든 결함이 있음을 나타냅니다.

Four configurations of a narrow tank sloshing problem were numerically simulated using a commercial CFD code designed for free surface flow. Two different fluids corresponding to Reynolds numbers of approximately 2 X 103 and 1 X 105 and two fluid depths were used to define the four cases. Mesh cell size independence tests were conducted for all four cases, but only a weak convergence towards the experimental results with increasing mesh resolution was found. The investigation also tested two different momentum advection numerical differencing schemes and found closer agreement using the 2nd order monotonicity preserving scheme than by using a first order scheme. Three turbulence models, including the default laminar flow, were tested but no preference was found for any model other than the laminar by virtue of its lower computational cost. Including the compressibility of the air-reduced agreement with the experimental data, the reasons for this are unclear.

The experimental pressure probe time history data sets included 100 values for each of the pressure peaks derived from 100 repeat tests, and thus we were able to examine the statistical significance of the agreement with the CFD simulations. Both the numerical simulations and the experiments gave rise to very steep pressure pulses corresponding to the sloshing wave impacts, and the experimental results were found to have a high degree of natural variability in the peak values. Although sensitivity tests of the CFD simulations (using, for example, slightly different initial starting conditions) were not formally conducted, the numerical solutions also showed a high degree of variability in the pressure peak magnitudes resulting from the use of different meshes, differencing schemes and turbulence models, which could be considered to show that the numerical solution also had a high degree of natural variability.

In all cases, the numerical solutions’ timing of the occurrence of the pressure peaks were found to be very accurate. However, even for the least challenging Lateral Oil case, the pressure peaks and the immediately following pressure values were overestimated, which indicated a shortcoming in the numerical modelling. When the experimental peak pressure variability was taken into account, the CFD-generated values fell inside the two Standard Deviation margin in nine of the 16 peaks rather than the 15 or more that would be required to show statistical significance in the sense that the CFD solution was representative of the experiment. This indicates that either the peaks are not normally distributed and/or the CFD model is in some way deficient at predicting them. Further work is required to establish how the peak pressures are distributed and/or to establish the physical reasons why the CFD model is overestimating the pressure peaks for even the least challenging Lateral Oil configuration.

References

  1. Spheric Benchmark Test Case, Sloshing Wave Impact Problem, Antonio Souto-Iglesias & Elkin Botia-Vera, https://wiki.manchester.ac.uk/spheric/index.php/Test10
  2. Peregrine DH (1993). Water-wave impact on walls. Annual Review of Fluid Mechanics. Vol 35, pp 23-43.

Editor’s Note

The complete document from which this note was extracted and the related data and input files are available on our Users Site. Readers are encouraged to read the original validation to get a full appreciation of the detail in this work investigating comparisons between simulation and experimental data. This study is especially noteworthy since it deals with highly non-linear sloshing of fluids interacting with the boundaries of a confining tank.

With regard to the author’s conclusions, it should be mentioned that the over prediction of fluid impact pressures in simulations could be the result of not allowing for sufficient compressibility effects in the liquids. For instance, in Fig. 3, it appears that there has been some air entrained in the liquid near the side wall. Also, negative pressures (i.e., below atmospheric) recorded experimentally might result from liquid drops remaining on the pressure sensors after the main body of liquid has drained away. Such details, which may be hard to quantify, only emphasize the difficulties involved in undertaking detailed validation studies. The author is commended for his excellent work.

Probabilistic investigation of cavitation occurrence in chute spillway based on the results of Flow-3D numerical modeling

Flow-3D 수치 모델링 결과를 기반으로 하는 슈트 여수로의 캐비테이션 발생 확률적 조사

Probabilistic investigation of cavitation occurrence in chute spillway based on the results of Flow-3D numerical modeling

Amin Hasanalipour Shahrabadi1*, Mehdi Azhdary Moghaddam2

1-University of Sistan and Baluchestan،amin.h.shahrabadi@gmail.com

2-University of Sistan and Baluchestan،Mazhdary@eng.usb.ac.ir

Abstract

Probabilistic designation is a powerful tool in hydraulic engineering. The uncertainty caused by random phenomenon in hydraulic design may be important. Uncertainty can be expressed in terms of probability density function, confidence interval, or statistical torques such as standard deviation or coefficient of variation of random parameters. Controlling cavitation occurrence is one of the most important factors in chute spillways designing due to the flow’s high velocity and the negative pressure (Azhdary Moghaddam & Hasanalipour Shahrabadi, ۲۰۲۰). By increasing dam’s height, overflow velocity increases on the weir and threats the structure and it may cause structural failure due to cavitation (Chanson, ۲۰۱۳). Cavitation occurs when the fluid pressure reaches its vapor pressure. Since high velocity and low pressure can cause cavitation, aeration has been recognized as one of the best ways to deal with cavitation (Pettersson, ۲۰۱۲). This study, considering the extracted results from the Flow-۳D numerical model of the chute spillway of Darian dam, investigates the probability of cavitation occurrence and examines its reliability. Hydraulic uncertainty in the design of this hydraulic structure can be attributed to the uncertainty of the hydraulic performance analysis. Therefore, knowing about the uncertainty characteristics of hydraulic engineering systems for assessing their reliability seems necessary (Yen et al., ۱۹۹۳). Hence, designation and operation of hydraulic engineering systems are always subject to uncertainties and probable failures. The reliability, ps, of a hydraulic engineering system is defined as the probability of safety in which the resistance, R, of the system exceeds the load, L, as follows (Chen, ۲۰۱۵): p_s=P(L≤R) (۱) Where P(۰) is probability. The failure probability, p_f, is a reliability complement and is expressed as follows: p_f=P[(L>R)]=۱- p_s (۲) Reliability development based on analytical methods of engineering applications has come in many references (Tung & Mays, ۱۹۸۰ and Yen & Tung, ۱۹۹۳). Therefore, based on reliability, in a control method, the probability of cavitation occurrence in the chute spillway can be investigated. In reliability analysis, the probabilistic calculations must be expressed in terms of a limited conditional function, W(X)=W(X_L ,X_R)as follows: p_s=P[W(X_L ,X_R)≥۰]= P[W(X)≥۰] (۳) Where X is the vector of basic random variables in load and resistance functions. In the reliability analysis, if W(X)> ۰, the system will be secure and in the W(X) <۰ system will fail. Accordingly, the eliability index, β, is used, which is defined as the ratio of the mean value, μ_W, to standard deviation, σ_W, the limited conditional function W(X) is defined as follows (Cornell, ۱۹۶۹): β=μ_W/σ_W (۴) The present study was carried out using the obtained results from the model developed by ۱:۵۰ scale plexiglass at the Water Research Institute of Iran. In this laboratory model, which consists of an inlet channel and a convergent thrower chute spillway, two aerators in the form of deflector were used at the intervals of ۲۱۱ and ۲۷۰ at the beginning of chute, in order to cope with cavitation phenomenon during the chute. An air duct was also used for air inlet on the left and right walls of the spillway. To measure the effective parameters in cavitation, seven discharges have been passed through spillway. As the pressure and average velocity are determined, the values of the cavitation index are calculated and compared with the values of the critical cavitation index, σ_cr. At any point when σ≤σ_cr, there is a danger of corrosion in that range (Chanson, ۱۹۹۳). In order to obtain uncertainty and calculate the reliability index of cavitation occurrence during a chute, it is needed to extract the limited conditional function. Therefore, for a constant flow between two points of flow, there would be the Bernoulli (energy) relation as follows (Falvey, ۱۹۹۰): σ= ( P_atm/γ- P_V/γ+h cos⁡θ )/(〖V_۰〗^۲/۲g) (۵) Where P_atm is the atmospheric pressure, γ is the unit weight of the water volume, θ is the angle of the ramp to the horizon, r is the curvature radius of the vertical arc, and h cos⁡θ is the flow depth perpendicular to the floor. Therefore, the limited conditional function can be written as follows: W(X)=(P_atm/γ- P_V/γ+h cos⁡θ )/(〖V_۰〗^۲/۲g) -σ_cr (۶) Flow-۳D is a powerful software in fluid dynamics. One of the major capabilities of this software is to model free-surface flows using finite volume method for hydraulic analysis. The spillway was modeled in three modes, without using aerator, ramp aerator, and ramp combination with aeration duct as detailed in Flow-۳D software. For each of the mentioned modes, seven discharges were tested. According to Equation (۶), velocity and pressure play a decisive and important role in the cavitation occurrence phenomenon. Therefore, the reliability should be evaluated with FORM (First Order Reliable Method) based on the probability distribution functions For this purpose, the most suitable probability distribution function of random variables of velocity and pressure on a laboratory model was extracted in different sections using Easy fit software. Probability distribution function is also considered normal for the other variables in the limited conditional function. These values are estimated for the constant gravity at altitudes of ۵۰۰ to ۷۰۰۰ m above the sea level for the unit weight, and vapor pressure at ۵ to ۳۵° C. For the critical cavitation index variable, the standard deviation is considered as ۰.۰۱. According to the conducted tests, for the velocity random variable, GEV (Generalized Extreme Value) distribution function, and for the pressure random variable, Burr (۴P) distribution function were presented as the best distribution function. The important point is to not follow the normal distribution above the random variables. Therefore, in order to evaluate the reliability with the FORM method, according to the above distributions, they should be converted into normal variables based on the existing methods. To this end, the non-normal distributions are transformed into the normal distribution by the method of Rackwitz and Fiiessler so that the value of the cumulative distribution function is equivalent to the original abnormal distribution at the design point of x_(i*). This point has the least distance from the origin in the standardized space of the boundary plane or the same limited conditional function. The reliability index will be equal to ۰.۴۲۰۴ before installing the aerator. As a result, reliability, p_s, and failure probability, p_f, are ۰.۶۶۲۹ and ۰.۳۳۷۱, respectively. This number indicates a high percentage for cavitation occurrence. Therefore, the use of aerator is inevitable to prevent imminent damage from cavitation. To deal with cavitation as planned in the laboratory, two aerators with listed specifications are embedded in a location where the cavitation index is critical. In order to analyze the reliability of cavitation occurrence after the aerator installation, the steps of the Hasofer-Lind algorithm are repeated. The modeling of ramps was performed separately in Flow-۳D software in order to compare the performance of aeration ducts as well as the probability of failure between aeration by ramp and the combination of ramps and aeration ducts. Installing an aerator in combination with a ramp and aerator duct greatly reduces the probability of cavitation occurrence. By installing aerator, the probability of cavitation occurrence will decrease in to about ۴ %. However, in the case of aeration only through the ramp, the risk of failure is equal to ۱۰%.

확률적 지정은 수력 공학에서 강력한 도구입니다. 유압 설계에서 임의 현상으로 인한 불확실성이 중요할 수 있습니다. 불확실성은 확률 밀도 함수, 신뢰 구간 또는 표준 편차 또는 무작위 매개변수의 변동 계수와 같은 통계적 토크로 표현될 수 있습니다. 캐비테이션 발생을 제어하는 ​​것은 흐름의 높은 속도와 음압으로 인해 슈트 여수로 설계에서 가장 중요한 요소 중 하나입니다(Azhdary Moghaddam & Hasanalipour Shahrabadi, ۲۰۲۰). 댐의 높이를 높이면 둑의 범람속도가 증가하여 구조물을 위협하고 캐비테이션으로 인한 구조물의 파손을 유발할 수 있다(Chanson, ۲۰۱۳). 캐비테이션은 유체 압력이 증기압에 도달할 때 발생합니다. 높은 속도와 낮은 압력은 캐비테이션을 유발할 수 있으므로, 통기는 캐비테이션을 처리하는 가장 좋은 방법 중 하나로 인식되어 왔습니다(Pettersson, ۲۰۱۲). 본 연구에서는 Darian 댐의 슈트 여수로의 Flow-۳D 수치모델에서 추출된 결과를 고려하여 캐비테이션 발생 확률을 조사하고 그 신뢰성을 조사하였다. 이 수력구조의 설계에서 수력학적 불확실성은 수력성능 해석의 불확실성에 기인할 수 있다. 따라서 신뢰성을 평가하기 위해서는 수력공학 시스템의 불확도 특성에 대한 지식이 필요해 보인다(Yen et al., ۱۹۹۳). 따라서 수력 공학 시스템의 지정 및 작동은 항상 불확실성과 가능한 고장의 영향을 받습니다. 유압 공학 시스템의 신뢰성 ps는 저항 R, 시스템의 부하 L은 다음과 같이 초과됩니다(Chen, ۲۰۱۵): p_s=P(L≤R)(۱) 여기서 P(۰)은 확률입니다. 고장 확률 p_f는 신뢰도 보완이며 다음과 같이 표현됩니다. Mays, ۱۹۸۰ 및 Yen & Tung, ۱۹۹۳). 따라서 신뢰성을 기반으로 제어 방법에서 슈트 여수로의 캐비테이션 발생 확률을 조사할 수 있습니다. 신뢰도 분석에서 확률적 계산은 제한된 조건부 함수 W(X)=W(X_L , X_R)은 다음과 같습니다. p_s=P[W(X_L,X_R)≥۰]= P[W(X)≥۰] (۳) 여기서 X는 부하 및 저항 함수의 기본 랜덤 변수 벡터입니다. 신뢰도 분석에서 W(X)> ۰이면 시스템은 안전하고 W(X) <۰에서는 시스템이 실패합니다. 따라서 표준편차 σ_W에 대한 평균값 μ_W의 비율로 정의되는 신뢰도 지수 β가 사용되며, 제한된 조건부 함수 W(X)는 다음과 같이 정의됩니다(Cornell, ۱۹۶۹). β= μ_W/σ_W (۴) 본 연구는 이란 물연구소의 ۱:۵۰ scale plexiglass로 개발된 모델로부터 얻은 결과를 이용하여 수행하였다. 이 실험 모델에서, 입구 수로와 수렴형 투수 슈트 여수로로 구성되며 슈트 중 캐비테이션 현상에 대처하기 위해 슈트 초기에 ۲۱۱과 ۲۷۰ 간격으로 편향기 형태의 2개의 에어레이터를 사용하였다. 여수로 좌우 벽의 공기 유입구에도 공기 덕트가 사용되었습니다. 캐비테이션의 효과적인 매개변수를 측정하기 위해 7번의 배출이 방수로를 통과했습니다. 압력과 평균 속도가 결정되면 캐비테이션 지수 값이 계산되고 임계 캐비테이션 지수 σ_cr 값과 비교됩니다. σ≤σ_cr일 때 그 범위에서 부식의 위험이 있다(Chanson, ۱۹۹۳). 슈트 중 캐비테이션 발생의 불확실성을 구하고 신뢰도 지수를 계산하기 위해서는 제한된 조건부 함수를 추출할 필요가 있다. 따라서 두 지점 사이의 일정한 흐름에 대해 다음과 같은 Bernoulli(에너지) 관계가 있습니다(Falvey, ۱۹۹۰). σ= ( P_atm/γ- P_V/γ+h cos⁡θ )/(〖V_۰〗 ^۲/۲g) (۵) 여기서 P_atm은 대기압, γ는 물의 단위 중량, θ는 수평선에 대한 경사로의 각도, r은 수직 호의 곡률 반경, h cos⁡ θ는 바닥에 수직인 흐름 깊이입니다. 따라서 제한된 조건부 함수는 다음과 같이 쓸 수 있습니다. W(X)=(P_atm/γ- P_V/γ+h cos⁡θ )/(〖V_۰〗^۲/۲g) -σ_cr (۶) Flow-۳D는 유체 역학의 강력한 소프트웨어. 이 소프트웨어의 주요 기능 중 하나는 수리학적 해석을 위해 유한 체적 방법을 사용하여 자유 표면 흐름을 모델링하는 것입니다. 방수로는 Flow-۳D 소프트웨어에 자세히 설명된 바와 같이 폭기 장치, 램프 폭기 장치 및 폭기 덕트가 있는 램프 조합을 사용하지 않고 세 가지 모드로 모델링되었습니다. 언급된 각 모드에 대해 7개의 방전이 테스트되었습니다. 식 (۶)에 따르면 속도와 압력은 캐비테이션 발생 현상에 결정적이고 중요한 역할을 합니다. 따라서 확률분포함수에 기반한 FORM(First Order Reliable Method)으로 신뢰도를 평가해야 한다 이를 위해 실험실 모델에 대한 속도와 압력의 확률변수 중 가장 적합한 확률분포함수를 Easy fit을 이용하여 구간별로 추출하였다. 소프트웨어. 확률 분포 함수는 제한된 조건부 함수의 다른 변수에 대해서도 정상으로 간주됩니다. 이 값은 단위 중량의 경우 해발 ۵۰۰ ~ ۷۰۰۰ m 고도에서의 일정한 중력과 ۵ ~ ۳۵ ° C에서의 증기압으로 추정됩니다. 임계 캐비테이션 지수 변수의 표준 편차는 ۰.۰۱으로 간주됩니다. . 수행된 시험에 따르면 속도 확률변수는 GEV(Generalized Extreme Value) 분포함수로, 압력변수는 Burr(۴P) 분포함수가 가장 좋은 분포함수로 제시되었다. 중요한 점은 확률 변수 위의 정규 분포를 따르지 않는 것입니다. 따라서 FORM 방법으로 신뢰도를 평가하기 위해서는 위의 분포에 따라 기존 방법을 기반으로 정규 변수로 변환해야 합니다. 이를 위해, 비정규분포를 Rackwitz와 Fiiessler의 방법에 의해 정규분포로 변환하여 누적분포함수의 값이 x_(i*)의 설계점에서 원래의 비정상분포와 같도록 한다. 이 점은 경계면의 표준화된 공간 또는 동일한 제한된 조건부 함수에서 원점으로부터 최소 거리를 갖습니다. 신뢰성 지수는 폭기 장치를 설치하기 전의 ۰.۴۲۰۴과 같습니다. 그 결과 신뢰도 p_s와 고장확률 p_f는 각각 ۰.۶۶۲۹과 ۰.۳۳۷۱이다. 이 숫자는 캐비테이션 발생의 높은 비율을 나타냅니다. 따라서 캐비테이션으로 인한 즉각적인 손상을 방지하기 위해 폭기 장치의 사용이 불가피합니다. 실험실에서 계획한 대로 캐비테이션을 처리하기 위해, 나열된 사양을 가진 두 개의 폭기 장치는 캐비테이션 지수가 중요한 위치에 내장되어 있습니다. 폭기장치 설치 후 캐비테이션 발생의 신뢰성을 분석하기 위해 Hasofer-Lind 알고리즘의 단계를 반복합니다. 경사로의 모델링은 폭기 덕트의 성능과 경사로에 의한 폭기 및 경사로와 폭기 덕트의 조합 사이의 실패 확률을 비교하기 위해 Flow-۳D 소프트웨어에서 별도로 수행되었습니다. 경사로 및 ​​폭기 덕트와 함께 폭기 장치를 설치하면 캐비테이션 발생 가능성이 크게 줄어듭니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 폭기 설치 후 캐비테이션 발생의 신뢰성을 분석하기 위해 Hasofer-Lind 알고리즘의 단계를 반복합니다. 경사로의 모델링은 폭기 덕트의 성능과 경사로에 의한 폭기 및 경사로와 폭기 덕트의 조합 사이의 실패 확률을 비교하기 위해 Flow-۳D 소프트웨어에서 별도로 수행되었습니다. 경사로 및 ​​폭기 덕트와 함께 폭기 장치를 설치하면 캐비테이션 발생 가능성이 크게 줄어듭니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 폭기장치 설치 후 캐비테이션 발생의 신뢰성을 분석하기 위해 Hasofer-Lind 알고리즘의 단계를 반복합니다. 경사로의 모델링은 폭기 덕트의 성능과 경사로에 의한 폭기 및 경사로와 폭기 덕트의 조합 사이의 실패 확률을 비교하기 위해 Flow-۳D 소프트웨어에서 별도로 수행되었습니다. 경사로 및 ​​폭기 덕트와 함께 폭기 장치를 설치하면 캐비테이션 발생 가능성이 크게 줄어듭니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 경사로의 모델링은 폭기 덕트의 성능과 경사로에 의한 폭기 및 경사로와 폭기 덕트의 조합 사이의 실패 확률을 비교하기 위해 Flow-۳D 소프트웨어에서 별도로 수행되었습니다. 경사로 및 ​​폭기 덕트와 함께 폭기 장치를 설치하면 캐비테이션 발생 가능성이 크게 줄어듭니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 경사로의 모델링은 폭기 덕트의 성능과 경사로에 의한 폭기 및 경사로와 폭기 덕트의 조합 사이의 실패 확률을 비교하기 위해 Flow-۳D 소프트웨어에서 별도로 수행되었습니다. 경사로 및 ​​폭기 덕트와 함께 폭기 장치를 설치하면 캐비테이션 발생 가능성이 크게 줄어듭니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다. 에어레이터를 설치하면 캐비테이션 발생 확률이 약 ۴%로 감소합니다. 그러나 램프를 통한 폭기의 경우 실패 위험은 ۱۰%와 같습니다.

Keywords

Aerator Probable Failure Reliability Method FORM Flow ۳D. 

electromagnetic metal casting computation designs Fig1

A survey of electromagnetic metal casting computation designs, present approaches, future possibilities, and practical issues

The European Physical Journal Plus volume 136, Article number: 704 (2021) Cite this article

Abstract

Electromagnetic metal casting (EMC) is a casting technique that uses electromagnetic energy to heat metal powders. It is a faster, cleaner, and less time-consuming operation. Solid metals create issues in electromagnetics since they reflect the electromagnetic radiation rather than consume it—electromagnetic energy processing results in sounded pieces with higher-ranking material properties and a more excellent microstructure solution. For the physical production of the electromagnetic casting process, knowledge of electromagnetic material interaction is critical. Even where the heated material is an excellent electromagnetic absorber, the total heating quality is sometimes insufficient. Numerical modelling works on finding the proper coupled effects between properties to bring out the most effective operation. The main parameters influencing the quality of output of the EMC process are: power dissipated per unit volume into the material, penetration depth of electromagnetics, complex magnetic permeability and complex dielectric permittivity. The contact mechanism and interference pattern also, in turn, determines the quality of the process. Only a few parameters, such as the environment’s temperature, the interference pattern, and the rate of metal solidification, can be controlled by AI models. Neural networks are used to achieve exact outcomes by stimulating the neurons in the human brain. Additive manufacturing (AM) is used to design mold and cores for metal casting. The models outperformed the traditional DFA optimization approach, which is susceptible to local minima. The system works only offline, so real-time analysis and corrections are not yet possible.

Korea Abstract

전자기 금속 주조 (EMC)는 전자기 에너지를 사용하여 금속 분말을 가열하는 주조 기술입니다. 더 빠르고 깨끗하며 시간이 덜 소요되는 작업입니다.

고체 금속은 전자기 복사를 소비하는 대신 반사하기 때문에 전자기학에서 문제를 일으킵니다. 전자기 에너지 처리는 더 높은 등급의 재료 특성과 더 우수한 미세 구조 솔루션을 가진 사운드 조각을 만듭니다.

전자기 주조 공정의 물리적 생산을 위해서는 전자기 물질 상호 작용에 대한 지식이 중요합니다. 가열된 물질이 우수한 전자기 흡수재인 경우에도 전체 가열 품질이 때때로 불충분합니다. 수치 모델링은 가장 효과적인 작업을 이끌어 내기 위해 속성 간의 적절한 결합 효과를 찾는데 사용됩니다.

EMC 공정의 출력 품질에 영향을 미치는 주요 매개 변수는 단위 부피당 재료로 분산되는 전력, 전자기의 침투 깊이, 복합 자기 투과성 및 복합 유전율입니다. 접촉 메커니즘과 간섭 패턴 또한 공정의 품질을 결정합니다. 환경 온도, 간섭 패턴 및 금속 응고 속도와 같은 몇 가지 매개 변수 만 AI 모델로 제어 할 수 있습니다.

신경망은 인간 뇌의 뉴런을 자극하여 정확한 결과를 얻기 위해 사용됩니다. 적층 제조 (AM)는 금속 주조용 몰드 및 코어를 설계하는 데 사용됩니다. 모델은 로컬 최소값에 영향을 받기 쉬운 기존 DFA 최적화 접근 방식을 능가했습니다. 이 시스템은 오프라인에서만 작동하므로 실시간 분석 및 수정은 아직 불가능합니다.

electromagnetic metal casting computation designs Fig1
electromagnetic metal casting computation designs Fig1
electromagnetic metal casting computation designs Fig2
electromagnetic metal casting computation designs Fig2
electromagnetic metal casting computation designs Fig3
electromagnetic metal casting computation designs Fig3
electromagnetic metal casting computation designs Fig4
electromagnetic metal casting computation designs Fig4
electromagnetic metal casting computation designs Fig5
electromagnetic metal casting computation designs Fig5
electromagnetic metal casting computation designs Fig6
electromagnetic metal casting computation designs Fig6
electromagnetic metal casting computation designs Fig7
electromagnetic metal casting computation designs Fig7
electromagnetic metal casting computation designs Fig8
electromagnetic metal casting computation designs Fig8
electromagnetic metal casting computation designs Fig9
electromagnetic metal casting computation designs Fig9

References

  1. 1.J. Sun, W. Wang, Q. Yue, Review on electromagnetic-matter interaction fundamentals and efficient electromagnetic-associated heating strategies. Materials 9(4), 231 (2016). https://doi.org/10.3390/ma9040231ADS Article Google Scholar 
  2. 2.E. Ghasali, A. Fazili, M. Alizadeh, K. Shirvanimoghaddam, T. Ebadzadeh, Evaluation of microstructure and mechanical properties of Al-TiC metal matrix composite prepared by conventional, electromagnetic and spark plasma sintering methods. Materials 10(11), 1255 (2017). https://doi.org/10.3390/ma10111255ADS Article Google Scholar 
  3. 3.D. Agrawal, Latest global developments in electromagnetic materials processing. Mater. Res. Innov. 14(1), 3–8 (2010). https://doi.org/10.1179/143307510×12599329342926Article Google Scholar 
  4. 4.S. Singh, P. Singh, D. Gupta, V. Jain, R. Kumar, S. Kaushal, Development and characterization of electromagnetic processed cast iron joint. Eng. Sci. Technol. Int. J. (2018). https://doi.org/10.1016/j.jestch.2018.10.012Article Google Scholar 
  5. 5.S. Singh, D. Gupta, V. Jain, Electromagnetic melting and processing of metal–ceramic composite castings. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 232(7), 1235–1243 (2016). https://doi.org/10.1177/0954405416666900Article Google Scholar 
  6. 6.S. Singh, D. Gupta, V. Jain, Novel electromagnetic composite casting process: theory, feasibility and characterization. Mater. Des. 111, 51–59 (2016). https://doi.org/10.1016/j.matdes.2016.08.071Article Google Scholar 
  7. 7.J. Lucas, J, What are electromagnetics? LiveScience. (2018). https://www.livescience.com/50259-Electromagnetics.html
  8. 8.R. Samyal, A.K. Bagha, R. Bedi, the casting of materials using electromagnetic energy: a review. Mater. Today Proc. (2020). https://doi.org/10.1016/j.matpr.2020.02.255Article Google Scholar 
  9. 9.S. Singh, D. Gupta, V. Jain, Processing of Ni-WC-8Co MMC casting through electromagnetic melting. Mater. Manuf. Process. (2017). https://doi.org/10.1080/10426914.2017.1291954Article Google Scholar 
  10. 10.R. Singh, S. Singh, V. Mahajan, Investigations for dimensional accuracy of investment casting process after cycle time reduction by advancements in shell moulding. Procedia Mater. Sci. 6, 859–865 (2014). https://doi.org/10.1016/j.mspro.2014.07.103Article Google Scholar 
  11. 11.R.R. Mishra, A.K. Sharma, On melting characteristics of bulk Al-7039 alloy during in-situ electromagnetic casting. Appl. Therm. Eng. 111, 660–675 (2017). https://doi.org/10.1016/j.applthermaleng.2016.09.122Article Google Scholar 
  12. 12.S. Zhang, 10 Different types of casting process. (2021). MachineMfg.com, https://www.machinemfg.com/types-of-casting/
  13. 13.Envirocare, Foundry health risks. (2013). https://envirocare.org/foundry-health-risks/
  14. 14.S.S. Gajmal, D.N. Raut, A review of opportunities and challenges in electromagnetic assisted casting. Recent Trends Product. Eng. 2(1) (2019)
  15. 15.R.R. Mishra, A.K. Sharma, Electromagnetic-material interaction phenomena: heating mechanisms, challenges and opportunities in material processing. Compos. Part A (2015). https://doi.org/10.1016/j.compositesa.2015.10.035Article Google Scholar 
  16. 16.S. Chandrasekaran, T. Basak, S. Ramanathan, Experimental and theoretical investigation on electromagnetic melting of metals. J. Mater. Process. Technol. 211(3), 482–487 (2011). https://doi.org/10.1016/j.jmatprotec.2010.11.001Article Google Scholar 
  17. 17.C.R. Bird, J.M. Mertz, U.S. Patent No. 4655276. (U.S. Patent and Trademark Office, Washington, DC, 1987)
  18. 18.R.R. Mishra, A.K. Sharma, Experimental investigation on in-situ electromagnetic casting of copper. IOP Conf. Ser. Mater. Sci. Eng. 346, 012052 (2018). https://doi.org/10.1088/1757-899x/346/1/012052Article Google Scholar 
  19. 19.V. Gangwar, S. Kumar, V. Singh, H. Singh, Effect of process parameters on hardness of AA-6063 in-situ electromagnetic casting by using taguchi method, in IOP Conference Series: Materials Science and Engineering, vol. 804(1) (IOP Publishing, 2020), p. 012019
  20. 20.X. Ye, S. Guo, L. Yang, J. Gao, J. Peng, T. Hu, L. Wang, M. Hou, Q. Luo, New utilization approach of electromagnetic thermal energy: preparation of metallic matrix diamond tool bit by electromagnetic hot-press sintering. J. Alloy. Compd. (2018). https://doi.org/10.1016/j.jallcom.2018.03.183Article Google Scholar 
  21. 21.S. Das, A.K. Mukhopadhyay, S. Datta, D. Basu, Prospects of Electromagnetic processing: an overview. Bull. Mater. Sci. 32(1), 1–13 (2009). https://doi.org/10.1007/s12034-009-0001-4Article Google Scholar 
  22. 22.K.L. Glass, D.M. Ashby, U.S. Patent No. 9050656. (U.S. Patent and Trademark Office, Washington, DC, 2015)
  23. 23.S. Verma, P. Gupta, S. Srivastava, S. Kumar, A. Anand, An overview: casting/melting of non ferrous metallic materials using domestic electromagnetic oven. J. Mater. Sci. Mech. Eng. 4(4), (2017). p-ISSN: 2393-9095; e-ISSN: 2393-9109
  24. 24.S.S. Panda, V. Singh, A. Upadhyaya, D. Agrawal, Sintering response of austenitic (316L) and ferritic (434L) stainless steel consolidated in conventional and electromagnetic furnaces. Scripta Mater. 54(12), 2179–2183 (2006). https://doi.org/10.1016/j.scriptamat.2006.02.034Article Google Scholar 
  25. 25.Y. Zhang, S. Yang, S. Wang, X. Liu, L. Li, Microwave/freeze casting assisted fabrication of carbon frameworks derived from embedded upholder in tremella for superior performance supercapacitors. Energy Storage Mater. (2018). https://doi.org/10.1016/j.ensm.2018.08.006Article Google Scholar 
  26. 26.D. Thomas, P. Abhilash, M.T. Sebastian, Casting and characterization of LiMgPO4 glass free LTCC tape for electromagnetic applications. J. Eur. Ceram. Soc. 33(1), 87–93 (2013). https://doi.org/10.1016/j.jeurceramsoc.2012.08.002Article Google Scholar 
  27. 27.M.H. Awida, N. Shah, B. Warren, E. Ripley, A.E. Fathy, Modeling of an industrial Electromagnetic furnace for metal casting applications. 2008 IEEE MTT-S Int. Electromagn. Symp. Digest. (2008). https://doi.org/10.1109/mwsym.2008.4633143Article Google Scholar 
  28. 28.P.K. Loharkar, A. Ingle, S. Jhavar, Parametric review of electromagnetic-based materials processing and its applications. J. Market. Res. 8(3), 3306–3326 (2019). https://doi.org/10.1016/j.jmrt.2019.04.004Article Google Scholar 
  29. 29.E.B. Ripley, J.A. Oberhaus, WWWeb search power page-melting and heat treating metals using electromagnetic heating-the potential of electromagnetic metal processing techniques for a wide variety of metals and alloys is. Ind. Heat. 72(5), 65–70 (2005)Google Scholar 
  30. 30.J. Campbell, Complete Casting Handbook: Metal Casting Processes, Metallurgy, Techniques and Design (Butterworth-Heinemann, 2015)Google Scholar 
  31. 31.B. Ravi, Metal Casting: Computer-Aided Design and Analysis, 1st edn. (PHI Learning Ltd, 2005)Google Scholar 
  32. 32.D.E. Clark, W.H. Sutton, Electromagnetic processing of materials. Annu. Rev. Mater. Sci. 26(1), 299–331 (1996)ADS Article Google Scholar 
  33. 33.A.D. Abdullin, New capabilities of software package ProCAST 2011 for modeling foundry operations. Metallurgist 56(5–6), 323–328 (2012). https://doi.org/10.1007/s11015-012-9578-8Article Google Scholar 
  34. 34.J. Ha, P. Cleary, V. Alguine, T. Nguyen, Simulation of die filling in gravity die casting using SPH and MAGMAsoft, in Proceedings of 2nd International Conference on CFD in Minerals & Process Industries (1999) pp. 423–428
  35. 35.M. Sirviö, M. Woś, Casting directly from a computer model by using advanced simulation software FLOW-3D Cast Ž. Arch. Foundry Eng. 9(1), 79–82 (2009)Google Scholar 
  36. 36.NOVACAST Systems, Nova-Solid/Flow Brochure, NOVACAST, Ronneby (2015)
  37. 37.AutoCAST-X1 Brochure, 3D Foundry Tech, Mumbai
  38. 38.EKK, Inc. Metal Casting Simulation Software and Consulting Services, CAPCAST Brochure
  39. 39.P. Muenprasertdee, Solidification modeling of iron castings using SOLIDCast (2007)
  40. 40.CasCAE, CT-CasTest Inc. Oy, Kerava
  41. 41.E. Dominguez-Tortajada, J. Monzo-Cabrera, A. Diaz-Morcillo, Uniform electric field distribution in electromagnetic heating applicators by means of genetic algorithms optimization of dielectric multilayer structures. IEEE Trans. Electromagn. Theory Tech. 55(1), 85–91 (2007). https://doi.org/10.1109/tmtt.2006.886913ADS Article Google Scholar 
  42. 42.B. Warren, M.H. Awida, A.E. Fathy, Electromagnetic heating of metals. IET Electromagn. Antennas Propag. 6(2), 196–205 (2012)Article Google Scholar 
  43. 43.S. Ashouri, M. Nili-Ahmadabadi, M. Moradi, M. Iranpour, Semi-solid microstructure evolution during reheating of aluminum A356 alloy deformed severely by ECAP. J. Alloy. Compd. 466(1–2), 67–72 (2008). https://doi.org/10.1016/j.jallcom.2007.11.010Article Google Scholar 
  44. 44.Penn State, Metal Parts Made In The Electromagnetic Oven. ScienceDaily. (1999) Retrieved May 8, 2021, from www.sciencedaily.com/releases/1999/06/990622055733.htm
  45. 45.R.R. Mishra, A.K. Sharma, A review of research trends in electromagnetic processing of metal-based materials and opportunities in electromagnetic metal casting. Crit. Rev. Solid State Mater. Sci. 41(3), 217–255 (2016). https://doi.org/10.1080/10408436.2016.1142421ADS Article Google Scholar 
  46. 46.D.K. Ghodgaonkar, V.V. Varadan, V.K. Varadan, Free-space measurement of complex permittivity and complex permeability of magnetic materials at Electromagnetic frequencies. IEEE Trans. Instrum. Meas. 39(2), 387–394 (1990). https://doi.org/10.1109/19.52520Article Google Scholar 
  47. 47.J. Baker-Jarvis, E.J. Vanzura, W.A. Kissick, Improved technique for determining complex permittivity with the transmission/reflection method. Microw. Theory Tech. IEEE Trans. 38, 1096–1103 (1990)ADS Article Google Scholar 
  48. 48.M. Bologna, A. Petri, B. Tellini, C. Zappacosta, Effective magnetic permeability measurementin composite resonator structures. Instrum. Meas. IEEE Trans. 59, 1200–1206 (2010)Article Google Scholar 
  49. 49.B. Ravi, G.L. Datta, Metal casting–back to future, in 52nd Indian Foundry Congress, (2004)
  50. 50.D. El Khaled, N. Novas, J.A. Gazquez, F. Manzano-Agugliaro. Microwave dielectric heating: applications on metals processing. Renew. Sustain. Energy Rev. 82, 2880–2892 (2018). https://doi.org/10.1016/j.rser.2017.10.043Article Google Scholar 
  51. 51.H. Sekiguchi, Y. Mori, Steam plasma reforming using Electromagnetic discharge. Thin Solid Films 435, 44–48 (2003)ADS Article Google Scholar 
  52. 52.J. Sun, W. Wang, C. Zhao, Y. Zhang, C. Ma, Q. Yue, Study on the coupled effect of wave absorption and metal discharge generation under electromagnetic irradiation. Ind. Eng. Chem. Res. 53, 2042–2051 (2014)Article Google Scholar 
  53. 53.K.I. Rybakov, E.A. Olevsky, E.V. Krikun, Electromagnetic sintering: fundamentals and modeling. J. Am. Ceram. Soc. 96(4), 1003–1020 (2013). https://doi.org/10.1111/jace.12278Article Google Scholar 
  54. 54.A.K. Shukla, A. Mondal, A. Upadhyaya, Numerical modeling of electromagnetic heating. Sci. Sinter. 42(1), 99–124 (2010)Article Google Scholar 
  55. 55.M. Chiumenti, C. Agelet de Saracibar, M. Cervera, On the numerical modeling of the thermomechanical contact for metal casting analysis. J. Heat Transf. 130(6), (2008). https://doi.org/10.1115/1.2897923Article MATH Google Scholar 
  56. 56.B. Ravi, Metal Casting: Computer-Aided Design and Analysis. (PHI Learning Pvt. Ltd., 2005)
  57. 57.J.H. Lee, S.D. Noh, H.-J. Kim, Y.-S. Kang, Implementation of cyber-physical production systems for quality prediction and operation control in metal casting. Sensors 18, 1428 (2018). https://doi.org/10.3390/s18051428ADS Article Google Scholar 
  58. 58.B. Aksoy, M. Koru, Estimation of casting mold interfacial heat transfer coefficient in pressure die casting process by artificial intelligence methods. Arab. J. Sci. Eng. 45, 8969–8980 (2020). https://doi.org/10.1007/s13369-020-04648-7Article Google Scholar 
  59. 59.S.S. Miriyala, V.R. Subramanian, K. Mitra, TRANSFORM-ANN for online optimization of complex industrial processes: casting process as case study. Eur. J. Oper. Res. 264(1), 294–309 (2018). https://doi.org/10.1016/j.ejor.2017.05.026MathSciNet Article MATH Google Scholar 
  60. 60.J.K. Kittu, G.C.M. Patel, M. Parappagoudar, Modeling of pressure die casting process: an artificial intelligence approach. Int. J. Metalcast. (2015). https://doi.org/10.1007/s40962-015-0001-7Article Google Scholar 
  61. 61.W. Chen, B. Gutmann, C.O. Kappe, Characterization of electromagnetic-induced electric discharge phenomena in metal-solvent mixtures. ChemistryOpen 1, 39–48 (2012)Article Google Scholar 
  62. 62.J. Walker, A. Prokop, C. Lynagh, B. Vuksanovich, B. Conner, K. Rogers, J. Thiel, E. MacDonald, Real-time process monitoring of core shifts during metal casting with wireless sensing and 3D sand printing. Addit. Manuf. (2019). https://doi.org/10.1016/j.addma.2019.02.018Article Google Scholar 
  63. 63.G.C. Manjunath Patel, A.K. Shettigar, M.B. Parappagoudar, A systematic approach to model and optimize wear behaviour of castings produced by squeeze casting process. J. Manuf. Process. 32, 199–212 (2018). https://doi.org/10.1016/j.jmapro.2018.02.004Article Google Scholar 
  64. 64.G.C. Manjunath Patel, P. Krishna, M.B. Parappagoudar, An intelligent system for squeeze casting process—soft computing based approach. Int. J. Adv. Manuf. Technol. 86, 3051–3065 (2016). https://doi.org/10.1007/s00170-016-8416-8Article Google Scholar 
  65. 65.M. Ferguson, R. Ak, Y.T. Lee, K.H. Law, Automatic localization of casting defects with convolutional neural networks, in 2017 IEEE International Conference on Big Data (Big Data) (Boston, MA, USA, 2017), pp. 1726–1735. https://doi.org/10.1109/BigData.2017.8258115.
  66. 66.P.K.D.V. Yarlagadda, Prediction of die casting process parameters by using an artificial neural network model for zinc alloys. Int. J. Prod. Res. 38(1), 119–139 (2000). https://doi.org/10.1080/002075400189617Article MATH Google Scholar 
  67. 67.G.C. ManjunathPatel, A.K. Shettigar, P. Krishna, M.B. Parappagoudar, Back propagation genetic and recurrent neural network applications in modelling and analysis of squeeze casting process. Appl. Soft Comput. 59, 418–437 (2017). https://doi.org/10.1016/j.asoc.2017.06.018Article Google Scholar 
  68. 68.J. Zheng, Q. Wang, P. Zhao et al., Optimization of high-pressure die-casting process parameters using artificial neural network. Int. J. Adv. Manuf. Technol. 44, 667–674 (2009). https://doi.org/10.1007/s00170-008-1886-6Article Google Scholar 
  69. 69.E. Mares, J. Sokolowski, Artificial intelligence-based control system for the analysis of metal casting properties. J. Achiev. Mater. Manuf. Eng. 40, 149–154 (2010)Google Scholar 
  70. 70.K.S. Senthil, S. Muthukumaran, C. Chandrasekhar Reddy, Suitability of friction welding of tube to tube plate using an external tool process for different tube diameters—a study. Exp. Tech. 37(6), 8–14 (2013)Article Google Scholar 
  71. 71.N.K. Bhoi, H. Singh, S. Pratap, P.K. Jain, Electromagnetic material processing: a clean, green, and sustainable approach. Sustain. Eng. Prod. Manuf. Technol. (2019). https://doi.org/10.1016/b978-0-12-816564-5.00001-3Article Google Scholar 
  72. 72.K.S. Senthil, D.A. Daniel, An investigation of boiler grade tube and tube plate without block by using friction welding process. Mater. Today Proc. 5(2), 8567–8576 (2018)Article Google Scholar 
  73. 73.E. Hetmaniok, D. Słota, A. Zielonka, Restoration of the cooling conditions in a three-dimensional continuous casting process using artificial intelligence algorithms. Appl. Math. Modell. 39(16), 4797–4807 (2015). https://doi.org/10.1016/j.apm.2015.03.056Article MATH Google Scholar 
  74. 74.C.V. Kumar, S. Muthukumaran, A. Pradeep, S.S. Kumaran, Optimizational study of friction welding of steel tube to aluminum tube plate using an external tool process. Int. J. Mech. Mater. Eng. 6(2), 300–306 (2011)Google Scholar 
  75. 75.T. Adithiyaa, D. Chandramohan, T. Sathish, Optimal prediction of process parameters by GWO-KNN in stirring-squeeze casting of AA2219 reinforced metal matrix composites. Mater. Today Proc. 150, 1598 (2020). https://doi.org/10.1016/j.matpr.2019.10.051Article Google Scholar 
  76. 76.B.P. Pehrson, A.F. Moore (2014). U.S. Patent No. 8708031 (U.S. Patent and Trademark Office, Washington, DC, 2014)
  77. 77.Liu, J., & Rynerson, M. L. (2008). U.S. Patent No. 7,461,684. Washington, DC: U.S. Patent and Trademark Office.
  78. 78.K. Salonitis, B. Zeng, H.A. Mehrabi, M. Jolly, The challenges for energy efficient casting processes. Procedia CIRP 40, 24–29 (2016). https://doi.org/10.1016/j.procir.2016.01.043Article Google Scholar 
  79. 79.R.R. Mishra, A.K. Sharma, Effect of solidification environment on microstructure and indentation hardness of Al–Zn–Mg alloy casts developed using electromagnetic heating. Int. J. Metal Cast. 10, 1–13 (2017). https://doi.org/10.1007/s40962-017-0176-1Article Google Scholar 
  80. 80.R.R. Mishra, A.K. Sharma, Effect of susceptor and Mold material on microstructure of in-situ electromagnetic casts of Al–Zn–Mg alloy. Mater. Des. 131, 428–440 (2017). https://doi.org/10.1016/j.matdes.2017.06.038Article Google Scholar 
  81. 81.S. Kaushal, S. Bohra, D. Gupta, V. Jain, On processing and characterization of Cu–Mo-based castings through electromagnetic heating. Int. J. Metalcast. (2020). https://doi.org/10.1007/s40962-020-00481-8Article Google Scholar 
  82. 82.S. Nandwani, S. Vardhan, A.K. Bagha, A literature review on the exposure time of electromagnetic based welding of different materials. Mater. Today Proc. (2019). https://doi.org/10.1016/j.matpr.2019.10.056Article Google Scholar 
  83. 83.F.J.B. Brum, S.C. Amico, I. Vedana, J.A. Spim, Electromagnetic dewaxing applied to the investment casting process. J. Mater. Process. Technol. 209(7), 3166–3171 (2009). https://doi.org/10.1016/j.jmatprotec.2008.07.024Article Google Scholar 
  84. 84.M.P. Reddy, R.A. Shakoor, G. Parande, V. Manakari, F. Ubaid, A.M.A. Mohamed, M. Gupta, Enhanced performance of nano-sized SiC reinforced Al metal matrix nanocomposites synthesized through electromagnetic sintering and hot extrusion techniques. Prog. Nat. Sci. Mater. Int. 27(5), 606–614 (2017). https://doi.org/10.1016/j.pnsc.2017.08.015Article Google Scholar 
  85. 85.V.R. Kalamkar, K. Monkova, (Eds.), Advances in Mechanical Engineering. Lecture Notes in Mechanical Engineering. (2021) https://doi.org/10.1007/978-981-15-3639-7
  86. 86.V. Bist, A.K. Sharma, P. Kumar, Development and microstructural characterisations of the lead casting using electromagnetic technology. Manager’s J. Mech. Eng. 4(4), 6 (2014). https://doi.org/10.26634/jme.4.4.2840Article Google Scholar 
  87. 87.A. Sharma, A. Chouhan, L. Pavithran, U. Chadha, S.K. Selvaraj, Implementation of LSS framework in automotive component manufacturing: a review, current scenario and future directions. Mater Today: Proc. (2021). https://doi.org/10.1016/J.MATPR.2021.02.374Article Google Scholar 
Figure 1 - General diagram of the forehead and body of the concentrated

Laboratory and Numerical Study of Dynamics Salty Density Current in The Reservoirs

저수지의 동적 염분 흐름의 실험 및 수치해석적 연구

Authors

1 Water resource expert Khuzestan Water and Power Authority
2 shahid chamran univercity of ahwaz

Since the characteristics of density current is affected by different parameters, the effect of discharge rate changes, gradient and the concentration of density current on speed of the forehead  and also the speed distribution in density current’s body have been investigated by physical and three-dimensional mathematical model (Flow-3d) in this research. For these purposes, different tests in the form of salty density current were done with three inflow discharge rates (0.7, 1 and 1.3 liters per second) and three different slopes (0, 1 and 2.2 percent). As well as to evaluate the effect of density changes on the flow characteristics, the concentration of 10, 15 and 20 grams per liter were used. In order to measure the speed of the forehead, velocity distribution in the body and its changes with flow, density and different slopes, video camera and ultrasound profiler speedometer were used in this study. Then, forehead speed and velocity distribution in the current’s body were achieved using six different turbulence models which are available on the software of “Flow-3D”. Comparing the results of physical and mathematical model showed that Eddy turbulence model and laminar flow mode have better accuracy in relation to other turbulent models. It should be noted that Reynolds number on experiments are at the range of  2000-4000.

밀도 흐름의 특성은 서로 다른 파라미터에 의해 영향을 받기 때문에 방출 속도 변화, 구배 및 밀도 흐름의 농도가 수두 속도에 미치는 영향과 밀도 흐름의 볼륨 속도 분포도 물리적 및 3차원 수학 모델(Flow-3d)에 의해 조사되었습니다.

이러한 목적을 위해 세 가지 유입 배출 속도(초당 0.7, 1 및 1.3L)와 세 가지 다른 경사도(0, 1, 2.2%)로 염분 밀도 흐름 형태의 다른 테스트가 수행되었습니다.

밀도 변화가 흐름 특성에 미치는 영향을 평가하기 위해 리터당 10, 15, 20g의 농도를 사용했습니다. 이 연구에서는 수두의 속도를 측정하기 위해 체내의 속도 분포와 흐름, 밀도 및 다양한 기울기와 함께 변화된 속도, 비디오 카메라 및 초음파 프로파일러 속도계를 사용했습니다.

그런 다음, “Flow-3D” 소프트웨어에서 사용할 수 있는 6가지 난류 모델을 사용하여 현재 볼륨의 수두 속도와 속도 분포를 달성했습니다.

물리적 모델과 수학적 모델의 결과를 비교한 결과, 에디 난류 모델과 층류 모드가 다른 난류 모델과 비교하여 더 나은 정확도를 가지고 있다는 것을 보여주었습니다.

레이놀즈 실험 번호는 2000-4000 범위라는 점에 유의해야 합니다.

Figure 1 - General diagram of the forehead and body of the concentrated
Figure 1 – General diagram of the forehead and body of the concentrated
Figure 2 - Dimensional profile of velocity distribution in concentrated flow (Graph and Altinacar, 1662)
Figure 2 – Dimensional profile of velocity distribution in concentrated flow (Graph and Altinacar, 1662)
Figure 1 - Schematic drawing of the physical model used
Figure 1 – Schematic drawing of the physical model used
Figure 0 - Sample of the concentrated flow created in the laboratory (front and body of concentrated flow)
Figure 0 – Sample of the concentrated flow created in the laboratory (front and body of concentrated flow)
Figure 6 - Mixing intensity values against Richardson number and comparing it with the results of other researchers
Figure 6 – Mixing intensity values against Richardson number and comparing it with the results of other researchers

Reference

1- حقی آبی، ا. 1383. بررسی اثر شیب کف بر پروفیل سرعت جریان غلیظ رساله دکتری رشته سازه های آبی ، دانشکده مهندسی علوم آب، دانشگاه شهید چمران اهواز.

2- کاهه، م. قمشی، م. و س، ح، موسوی جهرمی، 1391. بررسی آزمایشگاهی سرعت پیشروی جریان غلیظ بر روی سطوح زبر. علوم و مهندسی آبیاری، 35(1): 101-110.

3- کشتکار، ش. ایوب زاده، س ع. و ب، فیروزآبادی، 1389 . بررسی پروفیل سرعت و غلظت جریان گل آلود با استفاده از مدل فیزیکی. پژوهش‌های آبخیزداری،87(2): 43-36.

4- کوتی، ف. کاشفی پور، س، م. و م قمشی، 1391. تجزیه و تحلیل پروفیل های سرعت در جریان غلیظ. مجله ی علوم و فنون کشاورزی و منابع طبیعی، علوم آب و خاک، 59: 29-15.

5- Altinakar, M.S., Graf, W.H. and , E.J, Hopfinger. 1990. Weakly depositing turbidity current on a small slope. Journal of Hydraulic Research. 28(1): 55-80.

6- Baas, J.H. McCaffrey, W.D. Haughton P.D.W. and C, Choux. 2005. Coupling between suspended sediment distribution and turbulence structure in a laboratory turbidity current. Journal of Geophysics Research, 110: 20-32.

7- Barahmand, N. and A, Shamsai. 2010. Experimental and theoretical study of density jumps on smooth and rough beds”. Lakes and Reservoirs: Research and Management, 15(4): 285-307.

8- Britter, R.E. and P, Linden. 1980.The motion of the front of a gravity current traveling down an incline. Journal of Fluid Mechanics, 99(3): 531- 543.

9- Buckee, C. Kneller, B. and J, Peakall. 2001. Turbulence structure in steady solute-driven gravity currents Blackwell Oxford pp, 173-188.

10- Choux, C.M.A. Baas, J.H. McCaffrey, W.D. and P.D.W, Haughton. 2005. Comparison of spatio–temporal evolution of experimental particulate gravity flows at two different initial concentrations based on velocity grain size and density data. Sedimentary Geology, 179: 49-69.

11- FathiMoghadam, M. TorabiPoudeh, H. Ghomshi, M. and M, Shafaei. 2008. The density current head velocity in expansion reaches. Lakes & Reservoirs: Research & Management, 13(1): 63-68.

12- Ghomeshi, M. 1995. Reservoir sedimentationmodeling. Ph.D. Thesis. University of Wollongong. Australia.

  1. Graf, W.H. and M, S, Altinakar. 1998. Fluvial Hydraulics, Flow and Transport Processes in Channels of Simple Geometry. John Wiley and Sons, Ltd, England.

14- Ieong, K, K. Mok, K,M. and H, Yeh. 2006. Fluctuation of the front propagation speed of developed gravity current. Journal of Hydrodynamics, 18(3): 351-355.

15- LaRocca, M. Adduce, C. Sciortino, G. And A, B, Pinzon. 2008. Experimental and numerical simulation of three-dimensional gravity currents on smooth and rough bottom. Physics of Fluids, 20, 106603.

16- McCaffrey, W, D. Choux, C, M. Baas, J, H. And P, D, W, Haughton. 2003. Spatio-temporal evolution of velocity structure concentration and grainsize stratification within experimental particulate gravity currents. Marine and Petroleum Geology. 20: 851-860.

17- Sequeiros, O, E. Spinewine, B. Beaubouef, R, T. Sun, T. Garcia, H. M., and G, Parker. 2010. Characteristics of Velocity and Excess Density Profiles of Saline Underflows and Turbidity Currents Flowing over a Mobile Bed”. Journal of Hydraulic Engineering, 136(7): 167-180.

18- Turner, J, S. 1973. Buoyancy Effects in Fluids. Cambridge University Press London, U.K, pp. 178-181.

19- Yu, W, S. Lee, H, Y. And M, S, Hsu. 2000. Experiments on deposition behavior of fine in a reservoir. Journal of Hydraulic Engineering, 126(12): 912-920.

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/).

planar representation (cross-section at tank centre).

Analysis of cryogenic propellant behaviour in microgravity and low thrust environments*

미세 중력 및 저 추력 환경에서 극저온 추진체 거동 분석

M.F. Fisher, G.R. Schmidt and J.J. Martin
NASA Marshall Space Flight Center, Huntsville, AL 35824, USA

Abstract

우주선 비행 작업 (예 : 엔진 재시동 및 유체 전달) 중 극저온 추진제의 동작과 반응을 이해하는 것은 추진체 설계에서 매우 중요한 측면입니다. 엔진 연소 전 적절한 안정과 임무의 모든 단계에서 효과적인 차량 제어를 보장하려면 유체 움직임 및 슬로시 증폭에 대한 정확한 예측이 필요합니다.

이러한 유형의 분석을 강화하기 위해 Marshall Space Flight Center (MSFC)는 최근 Flow Sciences Inc에서 개발 한 CFD 패키지인 FLOW-3D를 인수했습니다. 이 문서에서는 FLOW-3D 모델 예측을 MSFC 드롭 타워 테스트 데이터와 비교한 최근 검증에 대해 설명합니다. 테스트는 원래 Saturn S-IVB 단계 액체 수소 (LH 2) 탱크의 설계 및 성능 평가를 지원하기 위해 1960 년대에 수행되었지만, 데이터는 FLOW-3D 모델의 정확성을 검증하는데 유용한 것으로 입증되었습니다.

Understanding the behaviour and response of cryogenic propellants during spacecraft flight operations (e.g., engine restart and fluid transfer) is an extremely important aspect of vehicle design. Accurate predictions of fluid motion and slosh amplification are needed to ensure proper settling prior to engine burn and effective vehicle control throughout all phases of the mission. To augment analyses of this type, Marshall Space Flight Center (MSFC) recently acquired FLOW-3D, a CFD package developed by Flow Sciences Inc. This paper describes a recent validation in which FLOW-3D model predictions were compared with MSFC drop tower test data. Although the tests were originally conducted in the 1960s to support design and performance assessments of the Saturn S-IVB stage liquid hydrogen (LH 2) tank, the data have proven useful for verifying the accuracy of the FLOW3D model.

Keywords: space cryogenics; propellants; microgravity

planar representation (cross-section at tank centre).
planar representation (cross-section at tank centre).
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 1. Geometries and bed topography settings of the nine computational fluid dynamics (CFD) simulations with channel curvature (C) changed from 0.77 to 0

The Straightening of a River Meander Leads to Extensive Losses in Flow Complexity and Ecosystem Services

Abstract

하천 복원 노력을 지원하기 위해 우리는 하천 파괴 속도를 늦출 필요가 있습니다. 이 연구는 하천 곡률 보호를 위해 구불 구불 한 하천이 곧게 펴질 때 수리적 복잡성 손실에 대한 자세한 설명을 제공합니다.

전산 유체 역학 (CFD) 모델링을 사용하여 채널 곡률 (C)이 잘 확립된 사행 굽힘 (C = 0.77)에서 곡률이 없는 직선 채널 (C = 0)로 저하되는 9 개의 시뮬레이션에서 유동 역학의 차이를 문서화했습니다.

공변량을 제어하고 수리적 복잡성에 대한 손실률을 늦추기 위해 각 9 개 채널 구현은 동등한 베드 형태 지형을 가졌습니다. 분석된 수력학적 변수에는 흐름 표면 고도, 흐름 방향 및 횡단 단위 배출, 흐름 방향, 가로 방향 및 수직 방향의 유속, 베드 전단 응력, 흐름 함수 및 채널 베드에서의 수직 저 유량 유속 비율이 포함되었습니다.

수력 복잡성의 손실은 처음에 수로를 C = 0.77에서 C = 0.33 (즉, 수로의 반경이 수로 폭의 3 배임) 할 때 점차적으로 발생했으며, 추가 직선화는 수력 복잡성에 대한 급속한 손실을 초래했습니다.

다른 연구에서는 수리적 복잡성이 중요한 하천 서식지를 제공하고 생물 다양성과 양의 상관 관계가 있음을 보여주었습니다. 이 연구는 강을 풀 때 수력학적 복잡성이 점진적으로 사라졌다가 빠르게 사라지는 방법을 보여줍니다.

To assist river restoration efforts we need to slow the rate of river degradation. This study provides a detailed explanation of the hydraulic complexity loss when a meandering river is straightened in order to motivate the protection of river channel curvature. We used computational fluid dynamics (CFD) modeling to document the difference in flow dynamics in nine simulations with channel curvature (C) degrading from a well-established tight meander bend (C = 0.77) to a straight channel without curvature (C = 0). To control for covariates and slow the rate of loss to hydraulic complexity, each of the nine-channel realizations had equivalent bedform topography. The analyzed hydraulic variables included the flow surface elevation, streamwise and transverse unit discharge, flow velocity at streamwise, transverse, and vertical directions, bed shear stress, stream function, and the vertical hyporheic flux rates at the channel bed. The loss of hydraulic complexity occurred gradually when initially straightening the channel from C = 0.77 to C = 0.33 (i.e., the radius of the channel is three-times the channel width), and additional straightening incurred rapid losses to hydraulic complexity. Other studies have shown hydraulic complexity provides important riverine habitat and is positively correlated with biodiversity. This study demonstrates how hydraulic complexity can be gradually and then rapidly lost when unwinding a river, and hopefully will serve as a cautionary tale.

Figure 1. Geometries and bed topography settings of the nine computational fluid dynamics (CFD) simulations with channel curvature (C) changed from 0.77 to 0
Figure 1. Geometries and bed topography settings of the nine computational fluid dynamics (CFD) simulations with channel curvature (C) changed from 0.77 to 0
Figure 2. Flow surface elevation (h) normalized by H at C = 0.77, C = 0.33, and C = 0 conditions. n denotes the lateral coordination with n = 0 at channel center and B denotes the channel width.
Figure 2. Flow surface elevation (h) normalized by H at C = 0.77, C = 0.33, and C = 0 conditions. n denotes the lateral coordination with n = 0 at channel center and B denotes the channel width.
Figure 3. Normalized flow surface profiles for the nine simulations at the point bar apex 1.5 s/B. The insert plot shows the second order derivative of normalized flow surface elevation in the transverse direction, Fh00(n/B), which gives the convexity or concavity of the surface profile curves.
Figure 3. Normalized flow surface profiles for the nine simulations at the point bar apex 1.5 s/B. The insert plot shows the second order derivative of normalized flow surface elevation in the transverse direction, Fh00(n/B), which gives the convexity or concavity of the surface profile curves.
Figure 4. Streamwise unit discharge qs/UH for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 4. Streamwise unit discharge qs/UH for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 5. Transverse unit discharge qn/UH for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 5. Transverse unit discharge qn/UH for channel curvature C = 0.77, 0.33, and 0 conditions.

Reference : https://www.mdpi.com/2073-4441/12/6/1680

Figure 6. Transverse unit discharge averaged over the transverse direction. The inset shows the R2 of transverse unit discharge < qn/UH > between each curvature, C, and the straight channel condition (C = 0, R2 = 1); a lower R2 suggests greater hydraulic complexity for transverse unit discharge.
Figure 6. Transverse unit discharge averaged over the transverse direction. The inset shows the R2 of transverse unit discharge < qn/UH > between each curvature, C, and the straight channel condition (C = 0, R2 = 1); a lower R2 suggests greater hydraulic complexity for transverse unit discharge.
Figure 7. Normalized depth averaged streamwise velocity <vs>/U for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 7. Normalized depth averaged streamwise velocity /U for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 8. The first moment of normalized depth averaged streamwise velocity <vs>/U, which represents center of gravity of the streamwise flow distribution, along the channel. The inset shows the R2 of the first moment of <vs>/U between each curvature and the straight channel condition (C = 0, R2 = 1); a lower R2 suggests greater hydraulic complexity for the first moment of depth averaged streamwise velocity.
Figure 8. The first moment of normalized depth averaged streamwise velocity /U, which represents center of gravity of the streamwise flow distribution, along the channel. The inset shows the R2 of the first moment of /U between each curvature and the straight channel condition (C = 0, R2 = 1); a lower R2 suggests greater hydraulic complexity for the first moment of depth averaged streamwise velocity.
Figure 9. Distribution of river channel bed shear Cf for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 9. Distribution of river channel bed shear Cf for channel curvature C = 0.77, 0.33, and 0 conditions.
Figure 10. Normalized vertical hyporheic flux vzbed/U at 2 mm below sediment surface for channel curvature C = 0.77, 0.33, and 0 conditions. Positive indicates upwelling of groundwater into the river channel.
Figure 10. Normalized vertical hyporheic flux vzbed/U at 2 mm below sediment surface for channel curvature C = 0.77, 0.33, and 0 conditions. Positive indicates upwelling of groundwater into the river channel.
Figure 11. Normalized vertical velocity <vz>/U for channel curvature C = 0.77, 0.33, and 0 conditions, with positive values upward flows, negative values downward flows.
Figure 11. Normalized vertical velocity /U for channel curvature C = 0.77, 0.33, and 0 conditions, with positive values upward flows, negative values downward flows.
Figure 12. Transverse stream function distribution ψ/UBH reveals the secondary circulation of transverse flow cells rotating at the meander apex 1.5 s/B for channel curvature C = 0.77 (A), C = 0.33 (B), and C = 0 (C), with positive values representing clockwise rotation direction when facing upstream, and negative values representing counter-clockwise rotation when facing upstream.
Figure 12. Transverse stream function distribution ψ/UBH reveals the secondary circulation of transverse flow cells rotating at the meander apex 1.5 s/B for channel curvature C = 0.77 (A), C = 0.33 (B), and C = 0 (C), with positive values representing clockwise rotation direction when facing upstream, and negative values representing counter-clockwise rotation when facing upstream.

References

  1. Paper 422-H); U.S. Government Printing Office: Washington, DC, USA, 1966.
  2. Leopold, L.B.; Wolman, M.G. River meanders. Bull. Geol. Soc. Am. 196071, 769–793. [Google Scholar] [CrossRef]
  3. Wohl, E. Rivers in the Landscape; John Wiley & Sons: Hoboken, NJ, USA, 2020. [Google Scholar]
  4. Dietrich, W.E.; Smith, J.D. Influence of the point bar on flow through curved channels. Water Resour. Res. 198319, 1173–1192. [Google Scholar] [CrossRef]
  5. Harvey, J.W.; Bencala, K. The effects of streambed topography on surface-subsurface water exchange in mountains catchments. Water Resour. Res. 199329, 89–98. [Google Scholar] [CrossRef]
  6. Bridge, J.S. Rivers and Floodplains: Forms, Processes, and Sedimentary Record; John Wiley & Sons: Hoboken, NJ, USA, 2009. [Google Scholar]
  7. Schumm, S.A. Patterns of alluvial rivers. Annu. Rev. Earth Planet. Sci. 198513, 5–27. [Google Scholar] [CrossRef]
  8. Vermeulen, B.; Hoitink, A.J.F.; Labeur, R.J. Flow structure caused by a local cross-sectional area increase and curvature in a sharp river bend. J. Geophys. Res. Earth Surf. 2015120, 1771–1783. [Google Scholar] [CrossRef]
  9. Konsoer, K.M.; Rhoads, B.L.; Best, J.L.; Langendoen, E.J.; Abad, J.D.; Parsons, D.R.; Garcia, M.H. Three-dimensional flow structure and bed morphology in large elongate meander loops with different outer bank roughness characteristics. Water Resour. Res. 201652, 9621–9641. [Google Scholar] [CrossRef]
  10. Li, B.D.; Zhang, X.H.; Tang, H.S.; Tsubaki, R. Influence of deflection angles on flow behaviours in openchannel bends. J. Mt. Sci. 201815, 2292–2306. [Google Scholar] [CrossRef]
  11. Gualtieri, C.; Abdi, R.; Ianniruberto, M.; Filizola, N.; Endreny, T.A. A 3D analysis of spatial habitat metrics about the confluence of Negro and Solimões rivers, Brazil. Ecohydrology 202013, e2166. [Google Scholar] [CrossRef]
  12. Gualtieri, C.; Ianniruberto, M.; Filizola, N.; Santos, R.; Endreny, T. Hydraulic complexity at a large river confluence in the Amazon basin. Ecohydrology 201710, e1863. [Google Scholar] [CrossRef]
  13. Kozarek, J.; Hession, W.; Dolloff, C.; Diplas, P. Hydraulic complexity metrics for evaluating in-stream brook trout habitat. J. Hydraul. Eng. 2010136, 1067–1076. [Google Scholar] [CrossRef]
  14. McCoy, E.D.; Bell, S.S.; Mushinsky, H.R. Habitat structure: Synthesis and perspectives. In Habitat Structure; Springer: Berlin, Germany, 1991; pp. 427–430. [Google Scholar]
  15. Re-Engineering Britain’s Rivers. The Economist. 6 March 2020. Available online: https://www.latestnigeriannews.com/news/8279579/reengineering-britains-rivers.html (accessed on 12 April 2020).
  16. Palmer, M.A.; Bernhardt, E.; Allan, J.; Lake, P.S.; Alexander, G.; Brooks, S.; Carr, J.; Clayton, S.; Dahm, C.; Follstad Shah, J.; et al. Standards for ecologically successful river restoration. J. Appl. Ecol. 200542, 208–217. [Google Scholar] [CrossRef]
  17. Abad, J.D.; Rhoads, B.L.; Güneralp, İ.; García, M.H. Flow structure at different stages in a meander-bend with bendway weirs. J. Hydraul. Eng. 2008134, 1052–1063. [Google Scholar] [CrossRef]
  18. Blanckaert, K.; Schnauder, I.; Sukhodolov, A.; van Balen, W.; Uijttewaal, W. Meandering: Field Experiments, Laboratory Experiments and Numerical Modeling. Technical Report. 2009. Available online: https://infoscience.epfl.ch/record/146621/files/2009-695-Blanckaert_et_al-Meandering_field_experiments_laboratory_experiments_and_numerical.pdf (accessed on 12 April 2020).
  19. Constantinescu, G.; Koken, M.; Zeng, J. The structure of turbulent flow in an open channel bend of strong curvature with deformed bed: Insight provided by detached eddy simulation. Water Resour. Res. 201147. [Google Scholar] [CrossRef]
  20. Sawyer, A.H.; Bayani Cardenas, M.; Buttles, J. Hyporheic exchange due to channel-spanning logs. Water Resour. Res. 201147. [Google Scholar] [CrossRef]
  21. Zhou, T.; Endreny, T. Meander hydrodynamics initiated by river restoration deflectors. Hydrol. Process. 201226, 3378–3392. [Google Scholar] [CrossRef]
  22. Hirt, C.W.; Nichols, B.D. Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 198139, 201–225. [Google Scholar] [CrossRef]
  23. Van Balen, W.; Uijttewaal, W.; Blanckaert, K. Large-eddy simulation of a curved open-channel flow over topography. Phys. Fluids 201022, 075108. [Google Scholar] [CrossRef]
  24. Blanckaert, K. Topographic steering, flow recirculation, velocity redistribution, and bed topography in sharp meander bends. Water Resour. Res. 201046. [Google Scholar] [CrossRef]
  25. Zeng, J.; Constantinescu, G.; Blanckaert, K.; Weber, L. Flow and bathymetry in sharp open-channel bends: Experiments and predictions. Water Resour. Res. 200844. [Google Scholar] [CrossRef]
  26. Elliott, A.H.; Brooks, N.H. Transfer of nonsorbing solutes to a streambed with bed forms: Laboratory experiments. Water Resour. Res. 199733, 137–151. [Google Scholar] [CrossRef]
  27. Zhou, T.; Endreny, T.A. Reshaping of the hyporheic zone beneath river restoration structures: Flume and hydrodynamic experiments. Water Resour. Res. 201349, 5009–5020. [Google Scholar] [CrossRef]
  28. Lane, S.; Bradbrook, K.; Richards, K.; Biron, P.; Roy, A. The application of computational fluid dynamics to natural river channels: Three-dimensional versus two-dimensional approaches. Geomorphology 199929, 1–20. [Google Scholar] [CrossRef]
  29. Vardy, A. Fluid Principles; McGraw-Hill International Series in Civil Engineering; McGraw-Hill: London, UK, 1990. [Google Scholar]
  30. Rozovskii, I.L. Flow of Water in Bends of Open Channels; Academy of Sciences of the Ukrainian SSR: Kiev, Ukraine, 1957. [Google Scholar]
  31. Blanckaert, K.; De Vriend, H.J. Secondary flow in sharp open-channel bends. J. Fluid Mech. 2004498, 353–380. [Google Scholar] [CrossRef]
  32. Johannesson, H.; Parker, G. Linear theory of river meanders. River Meand. 198912, 181–213. [Google Scholar] [CrossRef]
  33. Camporeale, C.; Perona, P.; Porporato, A.; Ridolfi, L. Hierarchy of models for meandering rivers and related morphodynamic processes. Rev. Geophys. 200745. [Google Scholar] [CrossRef]
  34. He, L. Distribution of primary and secondary currents in sine-generated bends. Water SA 201844, 118–129. [Google Scholar] [CrossRef]
  35. Liao, J.C.; Beal, D.N.; Lauder, G.V.; Triantafyllou, M.S. Fish exploiting vortices decrease muscle activity. Science 2003302, 1566–1569. [Google Scholar] [CrossRef]
  36. Crispell, J.K.; Endreny, T.A. Hyporheic exchange flow around constructed in-channel structures and implications for restoration design. Hydrol. Process. 20091168, 1158–1168. [Google Scholar] [CrossRef]
  37. Hester, E.T.; Gooseff, M.N. Moving Beyond the Banks: Hyporheic Restoration Is Fundamental to Restoring Ecological Services and Functions of Streams. Environ. Sci. Technol. 201044, 1521–1525. [Google Scholar] [CrossRef]
FIGURE 1. - FLOW-3D MODEL OF K-SITE TANK PRESSUR-IZATION.

Prediction of the Ullage Gas Thermal Stratification in a NASP Vehicle Propellant Tank Experimental Simulation Using FLOW-3D

FLOW-3D를 사용한 NASP 차량 추진 탱크 실험 시뮬레이션에서 Ullage 가스 열 층화 예측

Personal AuthorHardy, T. L.; Tomsik, T. M.

NASP (National Aero-Space Plane) 프로젝트의 일환으로 2D 온도 프로파일에 대한 중력, 초기 탱크 압력, 초기 유면 온도 및 열 전달 속도의 다차원 효과를 연구했습니다.

상업용 유한 차분 유체 흐름 모델인 FLOW-3D가 평가에 사용되었습니다. 이러한 효과는 기체 수소 가압제를 사용한 이전 액체 수소 실험 데이터를 기반으로 조사되었습니다.

FLOW-3D 결과는 기존 1D 모델과 비교되었습니다. 또한 메쉬 크기와 수렴 기준이 분석 결과에 미치는 영향을 조사했습니다. NASP 탱크 모델링을위한 향후 수정 및 FLOW-3D 사용에 대한 제안도 제공됩니다.

KeywordsAerospace planesComputer programsFluid flowHeat transferNational aerospace plane programPropellant tanksUllageComputational gridsConvergenceFinite difference theoryLiquid hydrogenMathematical modelsSimulationStratificationTemperature profiles
FIGURE 1. - FLOW-3D MODEL OF K-SITE TANK PRESSUR-IZATION.
FIGURE 1. – FLOW-3D MODEL OF K-SITE TANK PRESSUR-IZATION.
FIGURE 3. - EFFECT OF GRAVITY ON TEMPERATURE CONTOURS
FIGURE 3. – EFFECT OF GRAVITY ON TEMPERATURE CONTOURS
FIGURE 6.- EFFECT OF INITIAL PRESSURE ON VELOCITY PROFILE
FIGURE 6.- EFFECT OF INITIAL PRESSURE ON VELOCITY PROFILE
FIGURE 10. - EFFECT OF INITIAL TEMPERATURE ON TEMPERATURE CONTOURS
FIGURE 10. – EFFECT OF INITIAL TEMPERATURE ON TEMPERATURE CONTOURS
FIGURE 13. - EFFECT OF HEAT TRANSFER ON TEMPERATURE CONTOURS
FIGURE 13. – EFFECT OF HEAT TRANSFER ON TEMPERATURE CONTOURS
FIGURE 16. - EFFECT OF CONVERGENCE CRITERIA ON TEMPERATURE CONTOURS, 55 PERCENT ULLAGE, Pi = 17.4 PSI, 6 = 32.2 FT/SECZ, 24 SEC PRESSURIZATION
FIGURE 16. – EFFECT OF CONVERGENCE CRITERIA ON TEMPERATURE CONTOURS, 55 PERCENT ULLAGE, Pi = 17.4 PSI, 6 = 32.2 FT/SECZ, 24 SEC PRESSURIZATION
FIGURE 17. - COMPAR ISON OF CENTERLINE TEMPERATURES USING VARIOUS CONVERGENCE CRITERIA, 55 PERCENT ULLAGE, G = 32,2 FT/SEC2, P;= 17.4 PSI, 24 SEC PRESSURIZATION.
FIGURE 17. – COMPAR ISON OF CENTERLINE TEMPERATURES USING VARIOUS CONVERGENCE CRITERIA, 55 PERCENT ULLAGE, G = 32,2 FT/SEC2, P;= 17.4 PSI, 24 SEC PRESSURIZATION.
FIGURE 19. - EFFECT OF CONVERGENCE CRITERIA ON VELOCITY PROFILE,
FIGURE 19. – EFFECT OF CONVERGENCE CRITERIA ON VELOCITY PROFILE,
FIGURE 21. - EFFECT OF MESH SIZE ON TEMPERA- TURE CONTOURS, 55 PERCENT ULLAGE, Pi = 17.4 PSI, G = 0.0 FT/SEC2, , e = . 02, dt = , 005 SEC, 24 SEC PRESSURIZATION.
FIGURE 21. – EFFECT OF MESH SIZE ON TEMPERA- TURE CONTOURS, 55 PERCENT ULLAGE, Pi = 17.4 PSI, G = 0.0 FT/SEC2, , e = . 02, dt = , 005 SEC, 24 SEC PRESSURIZATION.

Figure 5.6 Experimental set-up equipped with high-speed camera system

COMPUTATIONAL FLUID DYNAMIC MODELLING OF LASER ADDITIVE MANUFACTURING PROCESS AND EFFECT OF GRAVITY

전산 유체 역학 레이저 첨가제 모델링 제조 공정 및 중력의 영향

A thesis submitted to
The University of Manchester
For the degree of
Doctor of Philosophy (PhD)
In the Faculty of Science and Engineering
2017
Heng Gu
School of Mechanical, Aerospace and Civil
Engineering

레이저 적층 제조 (LAM)는 재료를 층별로 선택적으로 추가하여 하나 또는 여러 개의 레이저 빔을 사용하여 재료를 융합하거나 응고시키는 3D 부품을 형성하는 것을 기반으로 합니다.

LAM 공정을 조사하는 데 상당한 양의 작업을 할 수 있지만 다른 재료 성장 방향에서 중력 및 동적 유체 흐름 특성의 영향에 대해서는 알려진 바가 거의 없습니다.

레이저 제조 기술의 발전과 함께 LAM은 실린더 본체, 터빈 블레이드의 표면 클래딩, 해양 드릴링 헤드, 다양한 증착 방향이 일반적으로 필요한 슬리브 및 몰드의 측벽을 비롯한 다양한 환경에서 점점 더 많이 사용되고 있습니다. 또한 공간 적층 제조의 경우 운영 환경이 매우 낮거나 무중력을 경험하게 됩니다.

LAM 프로세스를 모델링하기 위한 수치적 방법 개발에 대한 이전 연구에서 많은 노력을 기울였습니다. 그러나 이전 모델링 작업의 대부분은 자유 표면 형성을 고려하지 않고 용융 풀 역학 개발에 초점을 맞추었습니다. 몇 가지 조사에만 동적 유동 용융 풀에 대한 재료 추가 분석이 포함됩니다.

다양한 재료 증착 방향 및 무중력 효과에서 수행 할 때 모든 복잡한 기능을 사용하여 증착 프로세스를 시뮬레이션하고 중력 효과를 고려할 수 있는 모델을 개발하는 작업은 발견되지 않았습니다.

이 연구에서는 재료 추가, 표면 장력, 용융 및 응고, 중력, 온도 의존 재료 속성, 자유 표면 형성 및 이동을 포함한 복합 공정 요인을 고려한 LAM 공정을 위해 3 차원 과도 전산 유체 역학 모델이 ​​구축되었습니다. 열원. 레이저 금속 증착 공정에 대한 더 나은 이해는 수치적으로 그리고 실험적으로 이루어졌습니다.

이 연구는 단일 레이어의 증착, 여러 인접 패스 및 돌출 된 피쳐가 있는 완전한 3 차원 형상을 다루었습니다. 증착 공정 중 다양한 증착 방향과 무중력 및 매우 낮은 중력에 대한 중력의 영향을 조사하고 그 영향을 최소화하기 위해 공정 매개 변수를 최적화 했습니다.

이 연구는 또한 층별 재료 추가를 기반으로 레이저 좁은 갭 용접 공정의 기본 현상과 용접 공정이 다른 방향으로 수행 될 때 중력이 홈 내부의 용융 풀 형성에 미치는 영향을 이해하는 데까지 확장되었습니다.

용융 풀 개발 이력 및 온도 분포를 분석하여 공정 중에 표면 장력 계수의 영향을 논의했습니다. 현재 모델의 도움으로 증착 불균일성, 증착 양단의 돌출부, 경사, 융착 부족, 계단 효과, 표면 파형, 중력 변화로 인한 붕괴 등 다양한 결함을 설명 하였습니다.

이러한 모든 결함을 제거하기 위한 해당 솔루션이 제시되었습니다. 무중력 레이저 적층 제조에 대한 연구는 이전에 보고되지 않았던 몇 가지 새로운 현상을 발견하여 우주에서 미래의 레이저 3D 프린팅을 위한 길을 닦았습니다.

Figure 1.1 Diagram for thesis structure
Figure 1.1 Diagram for thesis structure
Figure 2.1 Basic construction of a laser system [8]
Figure 2.1 Basic construction of a laser system [8]
Figure 2.3 Schematic of a diode laser system [12]
Figure 2.3 Schematic of a diode laser system [12]
Figure 2.4 Principle of a cladding pumped fibre laser [13]
Figure 2.4 Principle of a cladding pumped fibre laser [13]
Figure 2.5 Concept of a thin disk laser [14]
Figure 2.5 Concept of a thin disk laser [14]
Figure 2.7 Lateral powder injection [12]
Figure 2.7 Lateral powder injection [12]
Figure 2.9 Laser additive manufacturing using wire, (a) front feeding, (b) rear feeding,  wire placed at (c) leading edge, (d) centre and (e) trailing edge of melt pool [23, 24]
Figure 2.9 Laser additive manufacturing using wire, (a) front feeding, (b) rear feeding, wire placed at (c) leading edge, (d) centre and (e) trailing edge of melt pool [23, 24]
Figure 2.20 Bead geometry at the beginning of the deposition with different surface  tension gradient (a) Negative, (b) positive, (c) Mixed [85]
Figure 2.20 Bead geometry at the beginning of the deposition with different surface tension gradient (a) Negative, (b) positive, (c) Mixed [85]
Figure 2.22 Simulation of humping effect in high-speed gas tungsten arc welding [91]
Figure 2.22 Simulation of humping effect in high-speed gas tungsten arc welding [91]
Figure 2.25 (a) Melt pool shape formed by Marangoni stress only, (b) Melt pool shape  formed by gravity force only, (c) Melt shape formed by the combination of those two  forces together [122]
Figure 2.25 (a) Melt pool shape formed by Marangoni stress only, (b) Melt pool shape formed by gravity force only, (c) Melt shape formed by the combination of those two forces together [122]
Figure 2.27 Growth rate and temperature gradient on solidification boundary with  different melt pool shape [120]
Figure 2.27 Growth rate and temperature gradient on solidification boundary with different melt pool shape [120]
Figure 2.29 Two different methods to produce overhang structures[136]
Figure 2.29 Two different methods to produce overhang structures[136]
Figure 2.30 Contact angle of a water droplet adhering on a glass window [142]
Figure 2.30 Contact angle of a water droplet adhering on a glass window [142]
Figure 2.31 Stress components of a single track laser deposition (a) x-direction, (b) ydirection, (c) z-direction, (d) von Mises equivalent stress [151]
Figure 2.31 Stress components of a single track laser deposition (a) x-direction, (b) ydirection, (c) z-direction, (d) von Mises equivalent stress [151]
Figure 2.32 Phase fraction of martensite during laser metal deposition [160]
Figure 2.32 Phase fraction of martensite during laser metal deposition [160]
Figure 4.15 Development of melt pool and velocity field 0.588 s, 1.2 s, 1.896 s, 2.4 s
Figure 4.15 Development of melt pool and velocity field 0.588 s, 1.2 s, 1.896 s, 2.4 s
Figure 4.33 Two methods to print C, (A) raster (B) offset out
Figure 4.33 Two methods to print C, (A) raster (B) offset out
Figure 5.4(a) Cavitar laser illumination system (b) High-speed camera in horizontal  position
Figure 5.4(a) Cavitar laser illumination system (b) High-speed camera in horizontal position
Figure 5.5 Schematic diagrams of wire laser deposition process (a) flat (b) vertical
Figure 5.5 Schematic diagrams of wire laser deposition process (a) flat (b) vertical
Figure 5.6 Experimental set-up equipped with high-speed camera system
Figure 5.6 Experimental set-up equipped with high-speed camera system
Figure 5.7 2-layer deposition result and cross-section (a) top view, (b) experimental  cross section, (c) cross-section of modelling result
Figure 5.7 2-layer deposition result and cross-section (a) top view, (b) experimental cross section, (c) cross-section of modelling result
Figure 5.13 Temperature and melt pool-velocity field history for case 8, (a&f:0.36 s,  b&g:1.44 s, c&h:1.80 s, d&i:1.908 s, e&j:2.196 s)
Figure 5.13 Temperature and melt pool-velocity field history for case 8, (a&f:0.36 s, b&g:1.44 s, c&h:1.80 s, d&i:1.908 s, e&j:2.196 s)
Figure 5.16 Comparison of melt pool evolution for cases with big and small spot size
Figure 5.16 Comparison of melt pool evolution for cases with big and small spot size
Figure 6.27 (a,b,c) before re-melting, (d,e,f) after re-melting
Figure 6.27 (a,b,c) before re-melting, (d,e,f) after re-melting

6.5 Conclusion

좁은 갭 용접 공정의 다양한 측면을 다루는 3 차원 모델이 구축되었습니다. 용접 비드와 측벽 사이의 융합 현상이 없는 것은 필러 재료와 측벽을 녹일 수 있는 충분한 에너지를 제공 할 수 없는 낮은 열 입력으로 인한 것일 수 있습니다.

증가된 레이저 출력을 적용하거나 재 용융 패스를 수행 한 후 더 나은 표면 품질을 얻을 수 있고 측벽과의 융합 부족을 제거 할 수 있습니다. 용접 비드의 모양이 볼록한 모양에서 오목한 모양으로 바뀌고 측면 벽과의 좋은 젖음이 실현 될 수 있습니다.

다양한 위치에서 좁은 틈새 용접에 대한 중력의 영향을 조사했습니다. 용융 풀 전면의 경사 모양은 중력의 영향으로 다르게 나타납니다.

반면, 홈이 없는 기판의 증착 공정과 비교할 때 대부분의 열을 전달하는데 도움이 되는 측벽의 존재로 인해 중력의 영향이 감소했습니다.

마지막 패스 중에 중력은 일부 평평하지 않은 위치에서 심각한 낙하 및 붕괴 문제를 일으킬 수 있습니다. 이것은 표면에 더 큰 용융 풀이 형성되어 중력과 표면 장력 사이의 균형이 깨졌기 때문입니다. 수직 업 위치에서 좁은 간격 용접 공정 동안 다른 중력 수준이 적용되었습니다.

용접 비드와 측벽 사이의 융합 부족은 중력 수준이 증가함에 따라 관찰 될 수 있습니다. 중력이 증가하면 용융 풀의 뒤쪽 영역으로 더 많은 액체 재료가 이동하여 더 심각한 물방울과 볼록한 모양의 용접 비드가 발생합니다.

용융 풀 개발 이력의 도움으로 용접 비드가 더 이상 그루브에 있지 않거나 측벽과의 직접적인 접촉이 적을 때 전도를 통해 더 적은 열이 방출 될 수 있기 때문에 용융 풀 부피가 크게 증가한다는 것을 알 수 있습니다.

좁은 간격 용접 공정에 대한 표면 장력 계수의 영향을 조사했습니다. 양의 표면 장력 계수를 적용하면 용접 비드가 홈 내부에서 덜 오목한 것처럼 보였고 측벽의 습윤 조건이 음의 ∂γ / ∂T 조건의 경우만큼 좋지 않았습니다.

측벽이 없으면 용접 비드는 표면의 마지막 패스 동안 음의 계수와 양의 계수 케이스 사이에 더 많은 차이를 보여줍니다. 표면 장력 계수는 홈 내부의 측벽과의 융합 상태를 결정하는 데 중요한 역할을 했습니다.

두꺼운 부분의 좁은 틈새 용접 중에 여러 번 통과하는 용접 비드 개발이 조사되었습니다. 비드 모양은 열 축적으로 인해 더 많은 패스가 증착 될수록 더 오목 해집니다. 패스 간의 융합 부족은 때때로 다음 패스의 재 용융 공정을 통해 제거 될 수 있습니다. 이종 재료를 사용한 좁은 틈새 용접 프로세스가 성공적으로 시뮬레이션되었습니다.

중심선을 따라 용융 풀과 용접 비드의 비대칭 형성은 재료 열 특성의 차이에 기인 할 수 있으며, 결과적으로 측벽과의 융합 부족을 유발할 수 있습니다.

비드 비대칭 문제는 수평 위치에서 용접 공정을 수행하거나 총 열 입력을 증가시켜 열전도율이 높은 측벽을 녹이는 방식으로 피할 수 있습니다. 재 용융 공정은 표면 품질을 향상시키고 모재와의 융착 문제를 제거하기 위해 용접된 표면에 적용 할 때 유용한 것으로 밝혀졌습니다.

圖1. 1 南海孤立內波空間分布圖(Hsu et al., 2000)

Numerical Modeling on Internal Solitary Wave propagation over an obstacle using Flow-3D

Keyword: Internal solitary waves, Numerical, Flow-3D, Computational Fluid Dynamics

연구자 : Yu-Ren Chen
지도교수 : Dr John R C Hsu
June 2012

기술과 수치 알고리즘의 발전으로 파도가 해양이나 항만 구조물에 미치는 영향에 대한 많은 연구가 개발되었으며,보다 정확한 결과를 얻기 위해 고효율 수치 계산 소프트웨어를 사용할 수 있습니다. 현재 내부 파 생성, 전송, 파동의 물리적 메커니즘은 국내외 해양 분야에서 중요한 연구 주제 중 하나입니다.

이 연구는 FLOW-3D 전산 유체 역학 (Computational Fluid Dynamics, CFD) 소프트웨어를 사용하여 상층의 담수와 하층의 담수를 시뮬레이션합니다. 바닷물의 밀도 계층화 유체는 중력 혼합 붕괴 방식을 사용하여 내부 파도를 생성하고 긴 경사와 같은 일반적인 장애물을 통해 파형 진화 및 유동장 분포를 탐구합니다.

짧은 플랫폼 사다리꼴 경사와 이등변 삼각형. 이 기사에서는 또한 소프트웨어 작동 설정과 FLOW-3D를 내부 파 실험에 적용하는 방법을 소개하고, 이전 실험 조건과 결과를 참조하여 내부 파 전송 과정을 시뮬레이션합니다. 시뮬레이션 결과는 실험 데이터를 확인하고 첫 번째 분석을 시뮬레이션합니다.

중력 붕괴 방식의 게이트의 개방 속도가 내부 파의 전송 시간 및 진폭에 미치는 영향; 시뮬레이션 결과는 게이트 개방 속도가 빠르고 내부 파의 진폭이 크고 전송 속도가 빠릅니다. ; 반대로 게이트 개방 속도가 느리면 내부 파의 진폭이 작고 전송 속도가 느리지 만 둘 다 비선형 비례 관계.

이 연구는 또한 다양한 장애물 (긴 기울기, 사다리꼴 기울기가있는 짧은 플랫폼, 이등변 삼각형)을 통한 내부 고독 파의 전송 과정을 시뮬레이션하고 단일 장애물을 통과하는 내부 파도의 파형 진화, 와류 및 유동장 변화를 논의합니다.

연구를 통해 우리가 매우 미세한 그리드를 사용하고 수치 시뮬레이션의 그래픽 출력을 열심히 분석 할 수 있다면 실험실 실험 관찰보다 내부 고독 파의 전송 특성을 더 잘 이해할 수 있다고 믿습니다.

요약

서로 다른 특성을 가진 두 유체의 계면에있는 파동을 계면 파라고합니다. 바다에서는 표층의 기압 변화에 의해 형성된 바람 장이 공기와 바다의 경계 파인 해면에 불어 올 때 변동을 일으킨다. 기체 또는 유체의 밀도 층화가 발생할 때 외부 힘 (예 : 바람, 압력, 파도 및 조류, 중력 등)에 의해 교란되면 내부 파도라고하는 경계면에서 변동이 발생할 수 있으므로 내부 파도가 발생할 수 있습니다. 웨이브는 밀도가 다른 층화 된 유체의 웨이브 현상입니다.

대기의 내부 파도와 같이 일상 생활에서 볼 수있는 내부 파도는 특히 오후 또는 비가 내리기 전에 깊고 얕은 altocumulus 구름 층으로 하늘에 자주 나타납니다. 대기 중의 내부 파의 움직임은 공기의 흐름에 영향을 주어 기류를 상승시키고 공기 중의 수증기가 물방울로 응축되어 구름이되도록합니다.

반대로 기류가 가라 앉으면 수증기가 응결이 쉽지 않습니다. 구름이 있어도 내부의 파도가 응결하기 어렵습니다. 소산되어 루버와 같은 altocumulus 구름을 형성합니다. 안정된 밀도와 층화 상태의 자연 수체는 외부 세계에 의해 교란 될 때 내부 파동 운동을 겪게됩니다.

예를 들어, 밀도가 안정되고 층화가 분명한 호수에서 바람 장은 수면에 파도에서 파생 된 내부 파동을 일으켜 물의 질량이 전달되고 호수 가장자리로 물이 축적되어 수위가 높아집니다. 위치 에너지를 형성하는 축적 영역; 수역이 가라 앉기 시작하면 위치 에너지를 운동 에너지로 변환하고 남미 콜롬비아의 Babine Lake의 내부 파동 거동과 같은 내부 파동 운동을 생성 할 수도 있습니다 (Farmer, 1978). ). 염분, 밀도 또는 온도가 안정된 바다에서는 조수와 지형의 영향으로 수역이 행성의 중력에 따라 움직입니다.

격렬한 기복이있는 지형을 통과 할 때 내부 파동이 발생합니다. ; 중국 해에서 발견되는 남쪽 내부 파도에서와 같이 (Hsu et al., 2000). 파동은 심해에서 얕은 물로 전달되며, 얕아 짐, 깨짐, 혼합, 소용돌이, 굴절, 회절 및 반사가있을 것입니다. 내부 파 전달은 일종의 파동이기 때문에 위에서 언급 한 파동 특성도 갖습니다.

해양 내부 파도는 길이가 수백 미터에서 수십 킬로미터에 이르는 광범위한 파장을 가지고 있으며,주기는 몇 분 정도 빠르며 수십 시간 정도 느리며 진폭은 몇 미터에서 수백 미터. 해양 내부 파도가 움직일 때 층화 위와 아래의 물 흐름 방향이 반대가되어 현재 전단 작용으로 인해 층화 경계면에서 큰 비틀림 힘이 발생합니다.

바다에 기초 말뚝과 같은 구조물이있는 경우 석유 시추 플랫폼의 고정 케이블은 큰 비틀림을 견딜 수 없어 파손될 가능성이 매우 높습니다 (Bole et al. 1994). 빽빽한 클라인 경계 근처에서 항해하는 잠수함이 해양 내부 파도 활동을 만나게되면 내부 파도에 의한 상승 전류로 인해 잠수함이 해저에 수면에 닿거나 충돌하여 잠수함이 손상 될 수 있습니다.

그러나 바다의 내부 파는 바람직하지 않으며 매우 중요한 역할을합니다. 예를 들어, 내부 파가 심해 지역에서 근해 대륙붕으로 전달되면 상하수 체가 교환됩니다. 해저에 영양분을 운반합니다. 선반 가장자리까지 생물학적 성장을 촉진하고 해당 지역의 생태 환경을 조절하며 (Osborne and Bruch et al., 1980; Sandstorm and Elliot et al., 1984) 어업 자원을 풍부하게합니다.

위에서 언급 한 항목 외에도 해저에 대한 케이블 및 파이프 라인, 수중 음파 탐지기, 해양 생물 환경, 군사 활동 등이 해양 내부 파도의 영향에 포함되므로 해양 내부 파도에 대한 연구가 매우 중요합니다.

최근 내부 파를 연구하는 방법에는 분석 이론 도출, 현장 조사 및 관찰, 실험실 실험 분석이 포함됩니다. 그러나 과학 기술의 급속한 발전, 발전과 발전, 컴퓨터의 대중화, 수치 계산 방법의 진화로 해양 공학과 관련된 많은 파동 효과는 일반적으로 수치 시뮬레이션 방법으로 해결됩니다.

또한 수치 연산 방법의 비용이 현장 조사 관측 및 실험실 실험 해석보다 저렴하고 시뮬레이션 결과를 더 빨리 얻을 수 있기 때문에 본 논문에서는 전산 유체 역학 (전산 유체 역학, 참조)의 FLOW-를 선정 하였다. 3D 소프트웨어는 내부 파 생성, 전송, 장애물 통과, 점차 소멸하는 움직임 과정을 시뮬레이션하고, 내부 파의 변화 과정을 분석하고 비교하기 위해 이전 실험실 모델 실험을 참조합니다.

圖1. 1  南海孤立內波空間分布圖(Hsu et al., 2000)
圖1. 1 南海孤立內波空間分布圖(Hsu et al., 2000)
圖1. 2  障礙高度與分層流體厚度關係之示意圖
圖1. 2 障礙高度與分層流體厚度關係之示意圖
圖3. 1 下沉型內孤立波通過梯形障礙的實驗配置圖(鄭明宏,2011)
圖3. 1 下沉型內孤立波通過梯形障礙的實驗配置圖(鄭明宏,2011)
圖3. 3  實驗室下沉型內孤立波經過13°斜坡梯形障礙物的連續組圖(鄭明宏,2011)
圖3. 3 實驗室下沉型內孤立波經過13°斜坡梯形障礙物的連續組圖(鄭明宏,2011)
圖3. 3 (a) 實驗室下沉型內孤立波(鄭明宏,2011;θ=13°,T = t0 = 42 s)
圖3. 3 (a) 實驗室下沉型內孤立波(鄭明宏,2011;θ=13°,T = t0 = 42 s)
圖3. 5 比較實驗室(上圖)內孤立波(圖3. 3 (a))與FLOW-3D模擬(下圖)的傳遞波形(θ=13°,t = 42 s)
圖3. 5 比較實驗室(上圖)內孤立波(圖3. 3 (a))與FLOW-3D模擬(下圖)的傳遞波形(θ=13°,t = 42 s)
圖4. 6閘門開啟速率0.14 m/s之等密度線及流場
圖4. 6閘門開啟速率0.14 m/s之等密度線及流場

圖4. 53 內波在三角形前坡反轉為順時針渦流,後坡面上形成逆時針渦流(t = 63 s)
圖4. 53 內波在三角形前坡反轉為順時針渦流,後坡面上形成逆時針渦流(t = 63 s)

Reference

Apel, J.R., Holbrook, J.R, Tsai, J. and Liu, A.K. (1985). The Sulu Sea internal soliton experiment. J. Phys. Oceanography, 15(12): 1625-1651. Ariyaratnam, J. (1998). Investigation of slope stability under internal wave action. B.Eng. (Hons.) thesis, Dept. of Environmental Eng., University of Western Australia, Australia. Baines, P.G. (1983). Tidal motion in submarine canyons – a laboratory experiment. J. Physical Oceanography, 13: 310-328. Benjamin, T.B. (1966). Internal waves of finite amplitude and permanent form. J. Fluid Mech., 25: 241-270. Bole, J.B., Ebbesmeyer, J.J. and Romea, R.D. (1994). Soliton currents in South China Sea: measurements and theoretical modelling. Proc. 26th Annual Offshore Tech. Conf., Houston, Texas. 367-375. Burnside, W. (1889). On the small wave-motions of a heterogeneous fluid under gravity. Proc. Lond., Math. Soc., (1) xx, 392-397. Chen C.Y., J.R-C. Hsu, H.H. Chen, C.F. Kuo and Cheng M.H (2007). Laboratory observations on internal solitary wave evolution on steep and inverse uniform slopes. Ocean Engineering, 34: 157-170. Cheng M.H., J.R-C. Hsu, C.Y. Chen (2005). Numerical model for internal solitary wave evolution on impermeable variable seabad, Proc.27th Ocean Eng, pp.355-359. Choi, W. and Camassa, R. (1996). Weakly nonlinear internal waves in a two-fluid system. J. Fluid Mech., 313: 83-103. Ebbesmeyer, C.C., and Romea, R.D. (1992). Final design parameters for solitons at selected locations in South China Sea. Final and supplementary reports prepared for Amoco Production Company, 209pp. plus appendices. Ekman, V. M., (1904). “On dead-water, Norwegian North Polar Expedition”, 1893-1896. Scientific Results, 5(15):1-150. Farmer, D.M. (1978). Observation of long nonlinear internal waves in a lake. J. Phys. Oceanography, 8(1): 63-73. Garret, C. and Munk, W. (1972). Space-time scales of internal waves. Geophys. Fluid Dyn., 3: 225-264. Gill, A.E. (1982). Atmosphere-Ocean Dynamics. International Geophysical Series, Vol. 30, San Diego, CA: Academic Press. Harleman, D.R.F. (1961). Stratified flow. Ch. 26 in Handbook of Fluid Dynamics (ed., V. Streeter), NY: McGraw-Hill, (26): 1-21. Helfrich, K.R. (1992). Internal solitary wave breaking and run-up on a uniform slope. J. Fluid Mech., 243: 133-154.

Helfrich, K.R. and Melville, W.K. (1986). On long nonlinear internal waves over slope-shelf topography. J. Fluid Mech., 167: 285-308. Honji, H., Matsunaga, N., Sugihara, Y. and Sakai, K. (1995). Experimental observation of interanl symmetric solitary waves in a two-layer fluid. Fluid Dynamics Research, 15 (2): 89-102. Hsu, M.K., Liu, A.K., and Liu, C. (2000). A study of internal waves in the China Sea and Yellow Sea using SAR. Continental Shelf Research, 20: 389-410. Johns, K. (1999). Interaction of an internal wave with a submerged sill in a two-layer fluid. B.Eng. (Hons.) thesis, Dept. of Environmental Eng., University of Western Australia, Australia Kao, T.W., Pan, F.S. and Renouard, D. (1985). Internal solitions on the pycnocline: generation, propagation, shoaling and breaking over a slope. J. Fluid Mech. 159: 19-53. Koop, C.G. and Butler, G. (1981). An investigation of internal solitary waves in a two-fluid system. J. Fluid Mech., 112: 225-251. Lin, T.W. (2001). A study on internal waves characteristics in north of South China Sea, Master Thesis, Institute of Oceanography, National Taiwan Univ., Taiwan. (In Chinese). Lynett, P., Wu, T.-R. and Liu, P. L.-F. (2002), Modeling wave runup with depth-integrated equations, Coastal Engineering, Vol. 46, pp. 89-107. Ming-Hung Cheng,John R.-C. Hsu, Chen-Yuan Chen and Cheng-Wu Chen (2009). Modelling the propagation of an internal solitary wave across double ridges and a shelf-slope.Environ Fluid Mech,9:321–340. Ming-Hung Cheng and John R.C. Hsu (2011). Effect of frontal slope on waveform evolution of a depression interfacial solitary wave across a trapezoidal obstacle. Ocean Engineering. Matsuno, Y. (1993). A unified theory of nonlinear wave propagation in two-layer fluid systems. J. Phys. Soc. Japan, 62: 1902-1916. Michallet, H. and Barthelemy, E. (1998). Experimental study of interfacial solitary waves. J. Fluid Mech., 366: 159-177. Muller, P. and X. Liu (2000). Scattering of internal waves at finite topography in two dimensions. Part I: Theory and case studies, J. Phys. Oceanogr., 30: 532-549 Nagashima, H. (1971). Reflection and breaking of internal waves on a sloping beach. J. Oceanographical Soc. Japan, 27(1): 1-6. Nansen, F. (1902). The oceanography of the north polar basin. Sci. Results, Norwegian North Polar Expedition 1893-1896, 3: 9. Osborne, A.R. and Burch, T.L. (1980). Internal solitons in the Andaman Sea. Science, 208 (43): 451-460

82 Russell, J.S. (1844). On waves. Report of the 14th Meeting of the British Association for the Advancement of Science, York, 311-390. Sandstrom, H. and Elliot J. A. (1984). Internal tide and solitons on the Scotian Shelf: a nutrient pump at work. Journal of Geophysical Research, 89 (C4): 6415-6428. Stokes G.G. (1847). On the Theory of Oscillatory Waves. Transactions of the Cambridge Philosophical Society, 8: 441–455. Strutt, J. W., Lord Rayleigh. (1883). Investigation of the character of the equilibrium of an incompressible heavy fluid of variable density.Proceedings of the London mathematical society, 8: pp. 170-177. Sveen, J.K., Guo, Y., Davies, P.A. and Grue, J. (2002). On the breaking of internal solitary waves at a ridge. J. Fluid Mech., 469 (25): 161-188. Vlasenko, V., and Hutter, K. (2002). Numerical experiments on the breaking of solitary internal waves over a slope-shelf topography. J. of Physical Oceanography, 32(6), pp.1779-1793. Wessels F. and Hutter K. (1996). Interaction of internal waves with a topographic sill in a two-layered fluid. J. Phys. Oceanogr , 26: 5-20

Figure 10.—Temperature contour time sequence for an EDS scale propellant tank at a jet mixing velocity of 0.06 m/s.

Computational Fluid Dynamics (CFD) Simulations of Jet Mixing in Tanks of Different Scales

NASA/TM—2010-216749

Kevin Breisacher and Jeffrey Moder
Glenn Research Center, Cleveland, Ohio

Prepared for the57th Joint Army-Navy-NASA-Air Force (JANNAF) Propulsion Meetingsponsored by the JANNAF Interagency Propulsion CommitteeColorado Springs, Colorado, May 3–7, 2010

Abstract

극저온 추진제의 장기 공간 저장을 위해 축류 제트 믹서는 탱크 압력을 제어하고 열 층화를 줄이기위한 하나의 개념입니다. 1960 년대부터 현재까지 10 피트 이하의 탱크 직경에 대한 광범위한 지상 테스트 데이터가 존재합니다.

Ares V EDS (Earth Departure Stage) LH2 탱크 용으로 계획된 것과 같이 직경이 30 피트 정도 인 탱크 용 축류 제트 믹서를 설계하려면 훨씬 더 작은 탱크에서 사용 가능한 실험 데이터를 확장하고 미세 중력을 설계해야 합니다.

이 연구는 10 배 차이가 나는 2 개의 탱크 크기에서 기존의 지상 기반 축류 제트 혼합 실험의 시뮬레이션을 수행하여 이러한 규모의 변화를 처리하는 전산 유체 역학 (CFD)의 능력을 평가합니다. 저궤도 (LEO) 해안 동안 Ares V 스케일 EDS LH2 탱크에 대한 여러 축 제트 구성의 시뮬레이션이 평가되고 선택된 결과도 제공됩니다.

두 가지 탱크 크기 (직경 1 및 10 피트)의 물을 사용하여 General Dynamics에서 1960 년대에 수행한 제트 혼합 실험 데이터를 사용하여 CFD 정확도를 평가합니다. 제트 노즐 직경은 직경 1 피트 탱크 실험의 경우 0.032 ~ 0.25 인치, 직경 10 피트 탱크 실험의 경우 0.625 ~ 0.875 인치였습니다.

제트 믹서를 켜기 전에 두 탱크에서 열 층화 층이 생성되었습니다. 제트 믹서 효율은 층화 층이 섞일 때까지 탱크의 열전대 레이크의 온도를 모니터링하여 결정되었습니다. 염료는 층화된 탱크에 자주 주입되었고 침투가 기록되었습니다. 실험 데이터에서 사용 가능한 속도나 난류량은 없었습니다.

제시된 시뮬레이션에는 자유 표면 추적 (Flow Science, Inc.의 FLOW-3D)이 포함된 시판되고 시간 정확도가 높은 다차원 CFD 코드가 사용됩니다. 서로 다른 시간에 탱크의 다양한 축 위치에서 계산 된 온도와 실험적으로 관찰된 온도를 비교합니다. 획득한 합의에 대한 다양한 모델링 매개 변수의 영향을 평가합니다.

Introduction

Constellation 프로그램의 일부인 Ares V는 우주 비행사를 달로 돌려 보내도록 설계된 무거운 리프트 발사기입니다. Ares V 스택의 일부인 EDS (Earth Departure Stage)는 지구의 중력에서 벗어나 승무원 차량과 달 착륙선을 달로 보내는데 필요합니다.

이러한 차량의 질량과 달로 보내는 데 필요한 에너지 때문에 EDS의 액체 수소(LH2)와 액체 산소(LO2) 추진제 탱크는 매우 클 것입니다(직경 10m). 탱크 내부로의 환경적 열 누출로 인해 혼합 장치를 포함한 열역학적 환기 시스템(TV)은 설계 한계 내에서 탱크 압력을 유지하고 엔진 시동에 필요한 한도 내에서 액체 온도를 유지하기 위해 며칠의 순서에 따라 공간 내 저장 기간 동안 필요할 수 있습니다.

이러한 혼합 장치 중 하나는 그림 1과 2와 같이 탱크 바닥 근처에 있는 (순가속과 관련하여) 탱크 축을 따라 중심에 있는 축 제트입니다. 축방향 제트 혼합기와 TVS에 통합된 것은 1960년대 중반부터 연구되어 왔으며(참조 1~5), 광범위한 축방향 제트 접지 테스트 데이터(비사이로젠(참조 1~9), 극저온(참조 10~16) 유체 사용), 에탄올을 사용한 일부 드롭 타워 테스트 데이터(참조 17 및 18)가 있습니다. 극저온 추진제를 사용하는 축방향 제트에 대한 기존 접지 테스트 데이터는 3m(10ft) 이하의 탱크 직경으로 제한됩니다.

저자가 알고 있는 바와 같이, 현재 임계 미달의 극저온 추진체를 사용하는 폐쇄형 탱크에 축방향 제트가 포함된 낙하탑, 항공기 또는 우주 비행 시험 데이터는 없습니다.

축방향 제트(Axial jet)는 지구 저궤도(LEO) 연안의 며칠 동안 EDS LH2 탱크에서 작동하는 혼합 장치의 후보 중 하나입니다. 제안된 EDS 탱크 척도의 극저온 저장 탱크에서 작동하는 축 제트 실험 데이터가 존재하지 않기 때문에, EDS 탱크를 위한 축 제트 TV의 초기 설계는 기존 데이터에 대해 고정된 상관 관계 및 CFD 분석에 의존할 필요가 있습니다.

이 연구는 두 개의 탱크 척도에서 크기 순서로 다른 축방향 제트 열분해 성능을 예측하기 위한 CFD 정확도 평가의 현재 진행 상황을 보고합니다. CFD 시뮬레이션은 물을 작동 유체로 사용하는 접지 테스트 축 제트 데이터(참조 1 – 4)와 비교됩니다. 이 평가를 위해 선택된 CFD 코드는 Flow Science(참조 21)의 상용 코드 FLOW-3D로, 극저온 저장 탱크 및 축방향 제트(참조 22~24)의 이전 분석에서 사용되었습니다.

LEO의 대표적인 EDS LH2 탱크에 대한 예비 축 제트 시뮬레이션도 여러 축 제트 구성에 대해 수행됩니다. 이러한 축방향 제트 구성의 열분해 성능을 평가하고 선택된 결과를 제시합니다.

이러한 예비 축방향 제트 EDS 시뮬레이션은 비교적 짧은 시간 동안 혼합기 성능만 평가합니다. 탱크 열 누출, 위상 변화 및 일반적인 자기 압력(제트 오프)/압력 붕괴(제트 온) 사이클을 포함한 보다 상세한 시뮬레이션이 향후 작업에서 추진될 수 있습니다.

Figure 1.—Schematic of the small water tank / Figure 2.—Schematic of the large water tank
Figure 1.—Schematic of the small water tank / Figure 2.—Schematic of the large water tank
Figure 5.—Temperature contours for large tank jet mixing simulation. (Temperature contour range 294 to 302 K)
Figure 5.—Temperature contours for large tank jet mixing simulation. (Temperature contour range 294 to 302 K)

상세 내용은 원문을 참조하시기 바랍니다.


Figure 9.—Schematic of a representative EDS scale propellant tank.
Figure 9.—Schematic of a representative EDS scale propellant tank.
Figure 10.—Temperature contour time sequence for an EDS scale propellant tank at a jet mixing velocity of 0.06 m/s.
Figure 10.—Temperature contour time sequence for an EDS scale propellant tank at a jet mixing velocity of 0.06 m/s.
Figure 14.—Temperature contour at t = 1000 s for the five jet mixer with a 0.06 m/s jet velocity
Figure 14.—Temperature contour at t = 1000 s for the five jet mixer with a 0.06 m/s jet velocity

Summary and Conclusions

사용 가능한 유사성 상관 관계를 사용하는 스케일링 전략은 EDS 클래스 제트 믹서에 대한 적절한 제트 크기 및 작동 조건을 결정하기 위해 개발되었습니다. 물 탱크 시뮬레이션에서 결정된 모델링 매개 변수를 사용하여 열 층화를 제어하기 위해 제트 믹서를 사용하여 EDS 등급 추진제 탱크의 혼합 이력에 대한 CFD 시뮬레이션을 수행했습니다.

시뮬레이션 결과는 다양한 믹싱 동작을 보여 주며 유사성 매개 변수의 사용에서 예상되는 것과 일치했습니다. 이러한 결과는 하위 규모 테스트 및 유사성 상관 관계와 함께 CFD 시뮬레이션이 EDS 등급 탱크를위한 효율적인 제트 믹서 설계를 허용 할 것이라는 확신을 제공합니다.

CFD 시뮬레이션은 다양한 크기의 직경과 제트를 가진 탱크의 제트 믹서에서 수행되었습니다. 1 피트 직경의 물 탱크에서 제트 혼합에 대해 사용 가능한 실험 데이터와 합리적으로 일치하는 모델링 매개 변수가 결정되었습니다. 동일한 모델링 매개 변수를 사용하여 대략 10 배 정도 떨어져있는 스케일로 워터 제트 혼합 실험에서 혼합을 시뮬레이션했습니다. 시뮬레이션 결과는 실험 온도 데이터와 잘 일치하는 것으로 나타났습니다.

References 1.Poth, L.J., Van Hook, J.R., Wheeler, D.M. and Kee, C.R., “A Study of Cryogenic Propellant Mixing Techniques. Volume 1 – Mixer design and experimental investigations,” NASA CR-73908, Nov 1968. 2.Poth, L.J., Van Hook, J.R., Wheeler, D.M. and Kee, C.R., “A Study of Cryogenic Propellant Mixing Techniques. Volume 2 – Experimental data Final report,” NASA CR-73909, Nov 1968. 3.Scale Experimental Mixing Investigations and Liquid-Oxygen Mixer Design,” NASA CR-113897, Sep 1970. 4.Van Hook, J.R. and Poth, L.J., “Study of Cryogenic Fluid Mixing Techniques. Volume 1 – Large-Van Hook, J.R., “Study of Cryogenic Fluid Mixing Techniques. Volume 2 – Large-Scale Mixing Data,” NASA CR-113914, Sep 1970. 5.Poth, L.J. and Van Hook, J.R., “Control of the Thermodynamic State of Space-Stored Cryogens by Jet Mixing,” J. Spacecraft, Vol. 9, No. 5, 1972. 6.Lovrich, T.N. and Schwartz, S.H., “Development of Thermal Stratification and Destratification Scaling Concepts – Volume II. Stratification Experimental Data,” NASA CR-143945, 1975. 7.Dominick, S.M., “Mixing Induced Condensation Inside Propellant Tanks,” AIAA–1984–0514. 8.Meserole, J.S., Jones, O.S., Brennan, S.M. and Fortini, A., “Mixing-Induced Ullage Condensation and Fluid Destratification,” AIAA–1987–2018. 9.Barsi, S., Kassemi, M., Panzarella, C.H. and Alexander, J.I., “A Tank Self-Pressurization Experiment Using a Model Fluid in Normal Gravity,” AIAA–2005–1143. 10.Stark, J.A. and Blatt, M.H., “Cryogenic Zero-Gravity Prototype Vent System,” NAS8-20146, Convair Report GDC-DDB67-006, Oct 1967. 11.Bullard, B.R., “Liquid Propellant Thermal Conditioning System Test Program,” NAS3-12033, Lockheed Missiles & Space Co., NASA CR-72971, July 1972. 12.Erickson, R.C., “Space LOX Vent System,” NAS8-26972, General Dynamics Convair Report CASD-NAS 75-021, April 1975.

13.Lin, C.S., Hasan, M.M. and Nyland, T.W., “Mixing and Transient Interface Condensation of a Liquid Hydrogen Tank,” NASA TM-106201 (or AIAA–1993–1968), 1993. 14.Lin, C.S., Hasan, M.M. and Van Dresar, N.T., “Experimental Investigation of Jet-Induced Mixing of a Large Liquid Hydrogen Storage Tank,” NASA TM-106629 (or AIAA–1994–2079), 1994. 15.Olsen, A.D., Cady, E.C., Jenkins, D.S. and Hastings, L., “Solar Thermal Upper Stage Cryogenic System Engineering Checkout Test,” AIAA–1999–2604. 16.Van Overbeke, T.J., “Thermodynamic Vent System Test in a Low Earth Orbit Simulation,” NASA/TM—2004-213193 (or AIAA–2004–3838), Oct 2004. 17.Aydelott, J.C., “Axial Jet Mixing of Ethanol in Cylindrical Containers During Weightlessness,” NASA-TP-1487, July 1979. 18.Aydelott, J.C., “Axial Modeling of Space Vehicle Propellant Mixing,” NASA-TP-2107, Jan 1983. 19.Bentz, M.D., “Tank Pressure Control in Low Gravity by Jet Mixing,” NASA CR–191012, Mar. 1993. 20.Hasan, M.M., Lin, C.S., Knoll, R.H. and Bentz, M.D., “Tank Pressure Control Experiment: Thermal Phenomena in Microgravity,” NASA-TP-3564, 1996. 21.FLOW-3D User’s Manual, version 9.4, Flow Science, Inc., Santa Fe, NM 2009. 22.Grayson, G.D., Lopez, A., Chandler, F.O., Hastings, L.J. and Tucker, S.P., “Cryogenic Tank Modeling for the Saturn AS-203 Experiment,” AIAA–2006–5258. 23.Lopez, A., Grayson, G.D., Chandler, F.O., Hastings, L.J., and Hedayat, A., “Cryogenic Pressure Control Modeling for Ellipsoidal Space Tanks,” AIAA–2007–5552. 24.Lopez, A., Grayson, G.D., Chandler, F.O., Hastings, L.J. and Hedayat, A., “Cryogenic Pressure Control Modeling for Ellipsoidal Space Tanks in Reduced Gravity,” AIAA–2008–5104. 25.Thomas, R.M., “Condensation of Steam on Water in Turbulent Motion,” Int. J. Multiphase Flow, Vol. 5, No. 1, pp. 1–15, 1979. 26.Zimmerli, G.A., Asipauskas, M., Chen, Y. and Weislogel, M.M., “A Study of Fluid Interface Configurations in Exploration Vehicle Propellant Tanks,” AIAA–2010–1294.

Simulation of EPS foam decomposition in the lost foam casting process

X.J. Liu a,∗, S.H. Bhavnani b,1, R.A. Overfelt c,2
a United States Steel Corporation, Great Lakes Works, #1 Quality Drive, Ecorse, MI 48229, United States b 213 Ross Hall, Department of Mechanical Engineering, Auburn University, Auburn, AL 36849-5341, United States c 202 Ross Hall, Department of Mechanical Engineering, Materials Engineering Program, Auburn University, Auburn, AL 36849-5341, United States
Received 17 April 2006; received in revised form 14 July 2006; accepted 21 August 2006

Keywords: Lost foam casting; Heat transfer coefficient; Gas pressure; VOF-FAVOR

LFC (Loss Foam Casting) 공정에서 부드러운 몰드 충진의 중요성은 오랫동안 인식되어 왔습니다. 충진 공정이 균일할수록 생산되는 주조 제품의 품질이 향상됩니다. 성공적인 컴퓨터 시뮬레이션은 금형 충전 공정에서 복잡한 메커니즘과 다양한 공정 매개 변수의 상호 작용을 더 잘 이해함으로써 새로운 주조 제품 설계의 시도 횟수를 줄이고 리드 타임을 줄이는데 도움이 될 수 있습니다.

이 연구에서는 용융 알루미늄의 유체 흐름과 금속과 발포 폴리스티렌 (EPS) 폼 패턴 사이의 계면 갭에 관련된 열 전달을 시뮬레이션하기 위해 전산 유체 역학 (CFD) 모델이 개발되었습니다.

상업용 코드 FLOW-3D는 VOF (Volume of Fluid) 방법으로 용융 금속의 전면을 추적 할 수 있고 FAVOR (Fractional Area / Volume Ratios) 방법으로 복잡한 부품을 모델링 할 수 있기 때문에 사용되었습니다. 이 코드는 폼 열화 및 코팅 투과성과 관련된 기체 갭 압력을 기반으로 다양한 계면 열 전달 계수 (VHTC)의 효과를 포함하도록 수정되었습니다.

수정은 실험 연구에 대해 검증되었으며 비교는 FLOW-3D의 기본 상수 열 전달 (CHTC) 모델보다 더 나은 일치를 보여주었습니다. 금속 전면 온도는 VHTC 모델에 의해 실험적 불확실성 내에서 예측되었습니다. 몰드 충전 패턴과 1-4 초의 충전 시간 차이는 여러 형상에 대해 CHTC 모델보다 VHTC 모델에 의해 더 정확하게 포착되었습니다. 이 연구는 전통적으로 매우 경험적인 분야에서 중요한 프로세스 및 설계 변수의 효과에 대한 추가 통찰력을 제공했습니다.

지난 20 년 동안 LFC (Loss Foam Casting) 공정은 코어가 필요없는 복잡한 부품을 제조하기 위해 널리 채택되었습니다. 이는 자동차 제조업체가 현재 LFC 기술을 사용하여 광범위한 엔진 블록과 실린더 헤드를 생산하기 때문에 알루미늄 주조 산업에서 특히 그렇습니다.

기본 절차, 적용 및 장점은 [1]에서 찾을 수 있습니다. LFC 프로세스는 주로 숙련 된 실무자의 경험적 지식을 기반으로 개발되었습니다. 발포 폴리스티렌 (EPS) 발포 분해의 수치 모델링은 최근에야 설계 및 공정 변수를 최적화하는 데 유용한 통찰력을 제공 할 수있는 지점에 도달했습니다. LFC 공정에서 원하는 모양의 발포 폴리스티렌 폼 패턴을 적절한 게이팅 시스템이있는 모래 주형에 배치합니다.

폼 패턴은 용융 금속 전면이 패턴으로 진행될 때 붕괴, 용융, 기화 및 열화를 겪습니다. 전진하는 금속 전면과 후퇴하는 폼 패턴 사이의 간격 인 운동 영역은 Warner et al. [2] LFC 프로세스를 모델링합니다. 금형 충진 과정에서 분해 산물은 운동 영역에서 코팅층을 통해 모래로 빠져 나갑니다.

용융 금속과 폼 패턴 사이의 복잡한 반응은 LFC 공정의 시뮬레이션을 극도로 어렵게 만듭니다. SOLA-VOF (SOLution AlgorithmVolume of Fluid) 방법이 Hirt와 Nichols [3]에 의해 처음 공식화 되었기 때문에 빈 금형을 사용한 전통적인 모래 주조 시뮬레이션은 광범위하게 연구되었습니다.

Lost foam 주조 공정은 기존의 모래 주조와 많은 특성을 공유하기 때문에이 새로운 공정을 모델링하는 데 적용된 이론과 기술은 대부분 기존의 모래 주조를 위해 개발 된 시뮬레이션 방법에서 비롯되었습니다. 패턴 분해 속도가 금속성 헤드와 금속 전면 온도의 선형 함수라고 가정함으로써 Wang et al. [4]는 기존의 모래 주조의 기존 컴퓨터 프로그램을 기반으로 복잡한 3D 형상에서 Lost foam 주조 공정을 시뮬레이션했습니다.

Liu et al. [5]는 금속 앞쪽 속도를 예측하기 위한 간단한 1D 수학적 모델과 함께 운동 영역의 배압을 포함했습니다. Mirbagheri et al. [6]은 SOLA-VOF 기술을 기반으로 금속 전면의 자유 표면에 대한 압력 보정 방식을 사용하는 Foam 열화 모델을 개발했습니다.

Kuo et al.에 의해 유사한 배압 방식이 채택되었습니다. [7] 운동량 방정식에서이 힘의 값은 실험 결과에 따라 패턴의 충전 순서를 연구하기 위해 조정되었습니다.

이러한 시뮬레이션의 대부분은 LFC 공정의 충전 속도가 기존의 모래 주조 공정보다 훨씬 느린 것으로 성공적으로 예측합니다. 그러나 Foam 분해의 역할은 대부분 모델의 일부가 아니며 시뮬레이션을 수행하려면 실험 데이터 또는 경험적 함수가 필요합니다.

현재 연구는 일정한 열전달 계수 (CHTC)를 사용하는 상용 코드 FLOW-3D의 기본 LFC 모델을 수정하여 Foam 열화와 관련된 기체 갭 압력에 따라 다양한 열전달 계수 (VHTC)의 영향을 포함합니다. 코팅 투과성. 수정은 여러 공정 변수에 대한 실험 연구에 대해 검증되었습니다.

또한, 손실 된 폼 주조에서 가장 중요한 문제인 결함 형성은 문헌에서 인용 된 수치 작업에서 모델링되지 않았습니다. 접힘, 내부 기공 및 표면 기포와 같은 열분해 결함은 LFC 작업에서 많은 양의 스크랩을 설명합니다. FLOW-3D의 결함 예측 기능은 프로세스를 이해하고 최적화하는데 매우 중요합니다.

Fig. 7. Comparison of mold filling times for a plate pattern with three ingates: (a) measured values by thermometric technique [18]; (b) predicted filling times based on basic CHTC model with gravity effect; and (c) predicted filing times based on the VHTC model with heat transfer coefficient changing with gas pressure; (d) mold filling time at the right-and wall of the mold for the plate pattern with three ingates.
Fig. 7. Comparison of mold filling times for a plate pattern with three ingates: (a) measured values by thermometric technique [18]; (b) predicted filling times based on basic CHTC model with gravity effect; and (c) predicted filing times based on the VHTC model with heat transfer coefficient changing with gas pressure; (d) mold filling time at the right-and wall of the mold for the plate pattern with three ingates.
Fig. 10. Defects formation predicted by (a) basic CHTC model with gravity effect; (b) VHTC model with heat transfer coefficient based on both gas pressure and coating thickness; and (c) improved model for two ingates. Color represents probability for defects (blue is the lowest and red highest).
Fig. 10. Defects formation predicted by (a) basic CHTC model with gravity effect; (b) VHTC model with heat transfer coefficient based on both gas pressure and coating thickness; and (c) improved model for two ingates. Color represents probability for defects (blue is the lowest and red highest).

References

[1] S. Shivkumar, L. Wang, D. Apelian, The lost-foam casting of aluminum alloy components, JOM 42 (11) (1990) 38–44.
[2] M.H. Warner, B.A. Miller, H.E. Littleton, Pattern pyrolysis defect reduction in lost foam castings, AFS Trans. 106 (1998) 777–785.
[3] C.W. Hirt, B.D. Nichols, Volume of Fluid (VOF) method for the dynamics of free boundaries, J. Comp. Phys. 39 (1) (1981) 201–225.
[4] C. Wang, A.J. Paul, W.W. Fincher, O.J. Huey, Computational analysis of fluid flow and heat transfer during the EPC process, AFS Trans. 101 (1993) 897–904.
[5] Y. Liu, S.I. Bakhtiyarov, R.A. Overfelt, Numerical modeling and experimental verification of mold filling and evolved gas pressure in lost foam casting process, J. Mater. Sci. 37 (14) (2002) 2997–3003.
[6] S.M.H. Mirbagheri, H. Esmaeileian, S. Serajzadeh, N. Varahram, P. Davami, Simulation of melt flow in coated mould cavity in the lost foam casting process, J. Mater. Process. Technol. 142 (2003) 493–507.
[7] J.-H. Kuo, J.-C. Chen, Y.-N. Pan, W.-S. Hwang, Mold filling analysis in lost foam casting process for aluminum alloys and its experimental validation, Mater. Trans. 44 (10) (2003) 2169–2174.
[8] C.W. Hirt, Flow-3D User’s Manual, Flow Science Inc., 2005.
[9] E.S. Duff, Fluid flow aspects of solidification modeling: simulation of low pressure die casting, The University of Queensland, Ph.D. Thesis, 1999.
[10] X.J. Liu, S.H. Bhavnani, R.A. Overfelt, The effects of foam density and metal velocity on the heat and mass transfer in the lost foam casting process, in: Proceedings of the ASME Summer Heat Transfer Conference, 2003,
pp. 317–323.
[11] W. Sun, P. Scarber Jr., H. Littleton, Validation and improvement of computer modeling of the lost foam casting process via real time X-ray technology, in: Multiphase Phenomena and CFD Modeling and Simulation in
Materials Processes, Minerals, Metals and Materials Society, 2004, pp. 245–251.
[12] T.V. Molibog, Modeling of metal/pattern replacement in the lost foam casting process, Materials Engineering, University of Alabama, Birmingham, Ph.D. Thesis, 2002.
[13] X.J. Liu, S.H. Bhavnani, R.A. Overfelt, Measurement of kinetic zone temperature and heat transfer coefficient in the lost foam casting process, ASME Int. Mech. Eng. Congr. (2004) 411–418.
[14] X. Yao, An experimental analysis of casting formation in the expendable
pattern casting (EPC) process, Department of Materials Science and Engineering, Worcester Polytechnic Institute, M.S. Thesis, 1994.
[15] M.R. Barkhudarov, C.W. Hirt, Tracking defects, Die Casting Engineer 43 (1) (1999) 44–52.
[16] C.W. Hirt, Modeling the Lost Foam Process with Defect PredictionsProgress Report: Lost-Foam Model Extensions, Wicking, Flow Science Inc., 1999.
[17] D. Wang, Thermophysical Properties, Solidification Design Center, Auburn University, 2001.
[18] S. Shivkumar, B. Gallois, Physico-chemical aspects of the full mold casting of aluminum alloys, part II: metal flow in simple patterns, AFS Trans. 95 (1987) 801–812.

Figure 1.2: Left panel: 3D CAD drawing of a printhead prototype showing (a) the melting unit, (b) the filter units, (c) the reservoir, (d) the static pressure hose, (e) the central part, and (f) the electronic driving supply. Image retrieved from [8]. Right panel: A schematic showing a single nozzle uint in the central part (e) of the printhead shown in the left panel.

Lattice Boltzmann method for contact line dynamics

접촉선 역학을 위한 Lattice Boltzmann 방법

ter verkrijging van de graad van doctor aan de
Technische Universiteit Eindhoven, op gezag van de
rector magnificus prof.dr.ir. C.J. van Duijn, voor een
commissie aangewezen door het College voor
Promoties, in het openbaar te verdedigen
op woensdag 7 mei 2014 om 16:00 uur

Introduction

움직이는 접촉선은 본질적으로 어디에나 존재하며, 표면에 미끄러지는 물방울은 우리가 일상에서 만나는 일반적인 예입니다. 유체 역학의 접촉선은 일반적으로 액체, 고체 및 주변 공기/증기 사이의 공통 경계라고합니다.

최근 미세 유체 공학의 발전으로 인해 접촉 라인의 역학을 제어하는 힘과 흐름 조건에 대한 근본적인 이해와 기술에 대한 많은 요구가 제기되었습니다. 이 논문은 접촉선의 물리학, 분석 및 수치 모델링 및 고무적인 산업 기하학과 관련된 측면을 포함합니다.

동기를 부여하는 산업 응용 분야는 이머전 리소그래피 (ASML)와 잉크젯 노즐 (Océ)의 프린트 헤드입니다. 이 두 가지 문제는 몇 가지 특징적인 길이 및 시간 척도, 고도로 구부러진 유체 인터페이스, 다상 흐름 및 복잡한 경계 조건을 포함하므로 분석 및 수치 연구가 어렵습니다.

포토 리소그래피는 서브 마이크론 정확도로 마스크에서 실리콘 웨이퍼로 패턴을 전송할 수 있는 복잡한 절차입니다 [1]. 포토 리소그래피 공정의 핵심 단계 중 하나는 고해상도 광학 시스템을 사용하여 실리콘 웨이퍼에 코팅 된 포토 레지스트를 DUV (심 자외선) 빛으로 노출시키는 것입니다. 광학 시스템을 사용하여 웨이퍼에 마스킹 할 수 있는 가장 작은 특징 또는 임계 치수 CD는 Rayleigh 기준으로 결정됩니다.

여기서 NA는 광학 시스템의 개구 수를 나타내고, λ는 사용 된 빛의 파장이고 k는 공정 종속 상수입니다. 광학 분야에서 광학 시스템의 개구 수 NA = n sin α는 시스템이 빛을 받아들이거나 방출 할 수 있는 각도 범위를 특성화하는 무차원 숫자입니다.

여기서 α는 렌즈의 수용 각도입니다 (0 < α <π / 2) 및 n은 렌즈와 포토 레지스트 사이의 매질의 굴절률입니다. CD의 가치가 감소하면 전자 장치가 더 작고 빨라집니다. 식에 의해 주어진 레일리 기준에 따르면. (1.1), 더 작은 CD 값은 k 또는 λ를 줄이거 나 NA를 늘림으로써 얻을 수 있습니다. 현재 KrF 및 ArF 엑시머 레이저의 경우 빛의 파장은 각각 최대 280nm 및 193nm까지 감소 될 수 있습니다 [1]. k는 분해능 향상 기술을 사용하여 0.4까지 감소 된 공정 의존 상수입니다 [2 ]. 개구 수는 sin α 또는 n을 증가시켜 증가시킬 수 있습니다.

sin α에 대한 실제 한계는 0.93으로, 이론적 한계 | sin α |에 매우 가깝습니다. ≤ 1. n을 늘리는 것이 이머전 리소그래피 사용의 기본 아이디어입니다. Immersion lithography는 렌즈와 포토 레지스트 사이의 에어 갭이 물로 대체되는 포토 리소그래피 기법입니다 (그림 1.1 (왼쪽 패널) 참조). 침지 리소그래피에 사용되는 물은 193nm 파장에 대해 1.44의 굴절률을 가진 고도로 정제 된 탈 이온수입니다 [3]. 이 굴절률 값은 분해 가능한 피처 크기의 해상도를 약 30 % 정도 증가시킵니다 [3].

이 방법은 훨씬 더 비싼 리소그래피 기술 [4]로 큰 변화를 가져 오지 않아도 된다는 장점을 가지고 더 작은 피처 크기를 달성하는 저렴한 방법입니다. 물이 웨이퍼의 포토 레지스트와 직접 접촉하기 때문에 이머전 리소그래피 기술은 주로 렌즈와 포토 레지스트의 오염 가능성과 관련된 몇 가지 문제를 야기합니다.

특히 웨이퍼 플레이트가 렌즈에 비해 Up 속도로 움직일 때 액체-공기-고체 접촉 라인도 움직입니다 (그림 1.1 (오른쪽 패널) 참조). 특정 최소 속도를 넘어 서면 전진 및 후퇴 접촉 선 (그림 1.1, 오른쪽 패널 참조)이 불안정 해지고 각각 공기를 동반하거나 액체 필름을 웨이퍼로 끌 수 있습니다 [5].

공기와 액체 필름은 결국 기포 나 액체 방울로 부서져서 리소그래피 공정에 부정적인 영향을 미칩니다. 이 논문에서 우리는 플레이트의 속도, 웨이퍼의 습윤 특성 및 주변 공기의 점도에 따라 전진 및 후퇴하는 접촉 라인의 안정성 연구에 기여했습니다.

1.1.2 Drop-on-demand inkjet printer

최신 잉크젯 인쇄 기술은 CIJ (연속 잉크젯) 및 DOD (주문형 드롭) 잉크젯의 두 가지 주요 유형으로 나눌 수 있습니다. CIJ 프린터에서 미세 노즐에서 나오는 액체 분사는 RP (Rayleigh-Plateau) 불안정성으로 인해 물방울로 분해됩니다. 이 RP 불안정성은 액체의 흐름을 정확하게 제어 할 수있는 음향 변동을 생성하는 압전 결정에 의해 유발되어 일정한 간격으로 물방울로 분해됩니다 [7].

DOD 잉크젯 프린터는 작동 원리에 따라 두 가지 범주로 더 나눌 수 있습니다 [8]. 여기서는 압전 잉크젯 (PIJ) 프린터에만 중점을 둡니다. PIJ 프린터에서 낙하 형성은 압전 소자에 의해 생성 된 압력 파에 의해 발생합니다. PIJ 프린터의 프린트 헤드 개략도가 그림 1.2에 나와 있습니다.

PIJ 프린터는 CIJ 프린터에 비해 상대적으로 느리지 만 인쇄 품질이 훨씬 더 높습니다 [7]. 프린터의 품질은 일반적으로 평방 인치당 도트 수 (dpi)로 측정되며 최신 응용 프로그램에는 더 작은 물방울 (높은 dpi)과 더 나은 정확도가 필요합니다. 방울의 정확도와 크기에 영향을 미치는 여러 요인 중에서 노즐, 노즐 플레이트의 젖음성 및 방울 형성 ​​빈도 fDOD가 중요한 역할을합니다 [8].

좋은 방울 형성을 위해 접촉 라인의 위치는 노즐 내에서 정밀하게 제어되어야 합니다. 이 논문에서는 PIJ 프린터에서 드롭 형성의 일부 측면에만 중점을 둡니다. 우리의 연구는 노즐 습윤성과 DOD 주파수가 방울 형성 ​​과정에 미치는 영향을 연구 할 수 있는 수치 도구의 개발을 목표로 합니다.

Figure 1.2: Left panel: 3D CAD drawing of a printhead prototype showing (a) the melting unit, (b) the filter units, (c) the reservoir, (d) the static pressure hose, (e) the central part, and (f) the electronic driving supply. Image retrieved from [8]. Right panel: A schematic showing a single nozzle uint in the central part (e) of the printhead shown in the left panel.
Figure 1.2: Left panel: 3D CAD drawing of a printhead prototype showing (a) the melting unit, (b) the filter units, (c) the reservoir, (d) the static pressure hose, (e) the central part, and (f) the electronic driving supply. Image retrieved from [8]. Right panel: A schematic showing a single nozzle uint in the central part (e) of the printhead shown in the left panel.
Figure 2.2: The liquid-vapor interface at the microscopic length scale obtained from a molecular dynamics (MD) simulation using Lennard-Jones potential
Figure 2.2: The liquid-vapor interface at the microscopic length scale obtained from a molecular dynamics (MD) simulation using Lennard-Jones potential. The vertical axis is in units of the molecular diameter σ and the stress shown in panel (c) is measured in /σ3 . Here,  is the energy scale corresponding to the intermolecular forces. (a) Snapshot of the liquid-vapor interface in the MD simulation. The red dotted line divides the system in two parts: Left and right. (b) Time-averaged normalized density profile ρ ∗ (z) across the interface. (c) Tangential force per unit area exerted by the left part on the right part of the system. The plot shows the difference between the normal and the tangential components of stress tensor: Π(z) = σ n − σ t . Images reproduced from [16].
Figure 2.3: Left panel: Water drops on a glass substrate
Figure 2.3: Left panel: Water drops on a glass substrate (Image source: http: // way2science. com/ molecular-theory-of-surface-tension).The red dotted line in the figure shows the position of the contact line. The shape of the big drops is affected by the force due to gravity. Right panel: Schematics of a liquid drop on a smooth non-deformable solid surface. The figure shows the contact angle, θe, in thermodynamic equilibrium.
Figure 6.1: Left panel: schematic of a single nozzle unit in the printhead
Figure 6.1: Left panel: schematic of a single nozzle unit in the printhead. Right panel: schematic of the channel-nozzle section of the printhead. The axisymmetric channel-nozzle section (right panel) is the simulation domain for our LB simulation (R = Rc).

FLOW-3D CAST

FLOW-3D CAST는 다양한 금속 주조 해석이 가능한 완벽한 열유동 해석 프로그램으로, 매우 정확한 모델링과 다기능성, 사용 용이성 및 고성능 클라우드 컴퓨팅 기능을 결합한 최첨단 금속 주조 해석 시뮬레이션 플랫폼입니다. 모든 금속 주조 공정에 대해 FLOW-3D CAST는  빠르고 직관적인 해석이 가능한 작업 공간을 제공합니다. 11개 공정에 대한 Workspace, 강력한 후처리, 충진 예측, 응고 및 결함 분석을 통해 FLOW-3D CAST는 최적의 주조 제품 설계에 필요한 도구와 로드맵을 모두 제공합니다.

FLOW-3D Cast는 거의 모든 주조 공정을 모델링 할 수 있도록 설계되었습니다. FLOW-3D Cast의 매우 정확한 유동 및 응고 결과는 표면 산화물, 혼입된 공기, 매크로 및 미세 다공성과 같은 중요한 주조 결함을 포착합니다. 다른 특별한 모델링 기능으로는 로봇 스프레이 냉각 및 윤활, 샷 슬리브 흐름 프로필, 스퀴즈 핀 및 열 응력을 모델링 할 수있는 열 다이 사이클링이 있습니다.

최적화된 시뮬레이션 설계를 통해 개발 시간을 단축하고 출시 시간을 단축하며 수율을 높일 수 있습니다. FLOW-3D CAST를 사용하면 설계 및 개발 비용을 절감할 수 있습니다.

FLOW-3D CAST Continuous Casting WorkspaceFLOW-3D CAST Gravity Die Casting Workspace
FLOW-3D CAST HPDC WorkspaceFLOW-3D CAST Investment Casting WorkspaceFLOW-3D CAST Low Pressure Sand Casting Workspace
FLOW-3D CAST Low Pressure Die Casting WorkspaceFLOW-3D CAST Sand Casting WorkspaceFLOW-3D CAST Sand Core Making Workspace
Lost Foam CastingFLOW-3D CAST Tilt Pour Casting
HPDC Oxides Simulation | FLOW-3D CAST
BMW Injector Casting Process – Innovative ingate system for gravity casting
Continuous Slab Casting | FLOW-3D CAST
Horizontal Centrifugal Pipe Casting | FLOW-3D CAST
Gravitational and sedimentation microfluidic technique (Huh et al. Anal Chem 2007)의 중력 회로도를 사용한 입자 분류

중력을 사용한 미세 유체 입자 분류

Microfluidics Particle Sorting Using Gravity

미세 유체 입자 분류는 진단, 화학적 및 생물학적 분석, 식품 및 화학 처리, 환경 평가에 적용됩니다. 이전 블로그에서 유체 역학을 사용한 미세 유체 입자 분류에 대해 이야기했습니다 . 같은 주제를 바탕으로 중력을 사용하여 미세 입자를 분류하는 또 다른 방법에 대해 논의하겠습니다. 아래 애니메이션에서 볼 수 있습니다.

유비쿼터스 중력(Ubiquitous gravity)은 미세 유체 장치에서 미세 입자를 분류하는 데 사용할 수 있습니다. 중력이 입자의 움직임에 수직으로 작용할 때 입자는 반경에 따른 속도로 안정됩니다. 또한 입자의 운동은 입자의 밀도, 유체의 밀도 및 유체의 점도 사이의 차이에서 비롯된 유체 역학적 효과의 영향을받습니다. 아래 이미지는 중력 분류 기술 회로도를 보여줍니다.

Gravitational and sedimentation microfluidic technique (Huh et al. Anal Chem 2007)의 중력 회로도를 사용한 입자 분류
Gravitational and sedimentation microfluidic technique (Huh et al. Anal Chem 2007)의 중력 회로도를 사용한 입자 분류

부력 대 항력

앞서 언급했듯이 중력은 서로 다른 입자가 서로 다른 속도로 침전되도록합니다. 모든 입자의 밀도가 같고 입자 밀도가 주변 유체의 밀도보다 낮 으면 부력 우세와 항력 우세라는 두 가지 유형의 분류를 사용할 수 있습니다. 반경이 더 큰 입자는 더 많은 부력을 경험하고 작은 입자 위의 경로를 따르는 경향이 있습니다. 그러나 외장 액체 (입자를 운반하는 용액)의 유입 속도가 충분히 높으면 항력 효과가 우세하기 시작하고 더 큰 입자가 더 작은 입자의 경로 아래로 이동하는 경향이 있습니다.

FLOW-3D 시뮬레이션 결과

경쟁하는 부력과 항력은 아래 FLOW-3D 에서 얻은 시뮬레이션 결과에서 명확하게 볼 수 있습니다 . 그림 1은 부력 지배적 인 입자 분류의 경우를 보여줍니다. 더 큰 (빨간색) 입자는 수평 채널의 상단을 향해 정렬됩니다. Fig. 2에 나타난 결과는 부력이 우세한 경우의 유입 초 속도를 20 배로 설정 한 후 얻은 것이다. 더 높은 입구 속도에서 더 큰 입자는 더 많은 운동량을 전달하므로 그 위치는 수직 부력의 영향을받지 않습니다. 따라서 입자는 수평 채널의 상단으로 올라가지 않습니다. 대신 그들은 계속해서 바닥으로 이동합니다.

부력

Buoyancy dominant sorting
Buoyancy dominant sorting

Drag

Figure 2. Drag dominant sorting
Figure 2. Drag dominant sorting

LOW-3D 의 입자 모델은입자 분류 또는 기타 입자 역학과 관련된 미세 유체 시뮬레이션에 성공적이고 쉽게 사용할 수 있습니다. 지금까지 우리는 FLOW-3D 의 입자 모델을사용하여 두 가지 입자 분류 기술을 보았습니다. 하나는 유체 역학을 사용하고 다른 하나는 중력을 사용합니다.

Figure 4 A view of the ogee spillway and Type 2 piers in the 3D CFD model

NUMERICAL ANALYSIS AND THE REAL WORLD : IT LOOKS PRETTY BUT IS IT RIGHT?

D. K. H. Ho, S. M. Donohoo, K. M. Boyes and C. C. Lock
Advanced Analysis, Worley Pty Limited
L7, 116 Miller Street, North Sydney, NSW 2060 Australia
Tel: +61 2 8923 6817 e-mail: david.ho@worley.com.au

Abstract

엔지니어링 설계에서 유한 요소, 유한 차분 및 전산 유체 역학 분석 소프트웨어와 같은 수치 도구의 일상적인 사용이 최근 몇 년 동안 증가했습니다. 소프트웨어 및 하드웨어 기술의 발전은보다 비선형적이고 복잡한 3 차원 분석이 수행되고 있음을 의미합니다.

그러나 본질적으로 “블랙 박스”인 이러한 강력한 소프트웨어는 “컴퓨팅”기술을 보유하고 있지만 광범위한 엔지니어링 경험이 필요하지 않은 분석가의 손에 “컴퓨터 보조 재해”로 이어질 수 있습니다. 품질 보증 절차의 엄격한 구현은 수치 모델이나 분석 기법이 정확한지 확인할 필요가 없을 수 있습니다.

이 백서에서는 복잡성이 증가하는 세 가지 실제 토목 공학 응용 프로그램에서 수치 분석 결과를 검증하는 방법을 설명합니다. 여기에는 유한 요소법을 이용한 수조 탱크의 구조 해석, 전산 유체 역학법을 이용한 수력 구조물 위의 홍수 조사, 유한 ​​차분법을 이용한 안벽 시공 시뮬레이션 등이 있습니다. 입력 데이터의 불확실성 수준과 각 사례에 대한 계산 결과의 신뢰성에 대해 논의합니다. 분석 과정에서 몇 가지 흥미로운 결과가 발견되었습니다.

첫 번째 사례 연구는 시공의 질이 구조물의 성능에 상당한 영향을 미친다는 것을 보여주었습니다. 그러나 설계자는 설계 단계에서 이러한 상황을 수량화하고 분석하지 못할 수도 있습니다. 필요할 경우 향후 역분석은 물론 설계 검증의 기준점이 될 수 있도록 공사 종료 시 모니터링의 중요성이 필수적입니다. 유한 요소 분석은 복잡한 문제를 분석할 수 있는 강력한 수치 도구이지만, 분석가들은 문제의 행동이 단순하고 잘 이해되는 것처럼 보일 수 있는 상황에서 예상치 못한 결과를 만날 수 있도록 준비해야 합니다.

두 번째 사례 연구에서는 중요한 배수로 구조에 전산 유체 역학 분석이 처음으로 적용 되었기 때문에 엄격한 검증 프로세스가 강조됩니다. 그것은 2D ogee 방수로 프로파일로 시작하여 문제의 방수로의 3D 모델을 분석하기 위해 진행되는 방식으로 수행되었습니다.
계산된 결과를 각 단계에서 이론 및 물리적 테스트 데이터와 비교했습니다. 유체 흐름 문제의 비선형적 특성에도 불구하고, 분석은 확신을 가지고 실제 설계 목적에 적합한 결과를 제공할 수 있었습니다.

최종 사례 연구에서는 안벽의 거동이 시공 이력과 매립 방식에 영향을 받은 것으로 나타났습니다. 벽의 움직임은 매우 가변적인 토양 속성에도 불구하고 질적으로도 단순한 비선형 토양 모델을 사용하여 정확하게 예측되었습니다. 지속적인 모니터링 기록이 없기 때문에 검증은 어려웠습니다. 계산된 결과를 검증하는 열쇠는 수치 소프트웨어 도구를 사용하지 않는 독립적인 계산을 찾는 것입니다. 대부분의 경우 이러한 솔루션을 사용할 수 있습니다. 그러나 다른 경우에는 실험실 또는 현장 관찰에만 의존할 수 있습니다.

Introduction

오늘날 수치 해석은 대부분의 엔지니어링 설계에서 필수적인 부분을 형성합니다. 따라서 결과 검증의 필요성은 분석 기술 / 방법론을 신뢰할 수 있고 설계자가 계산 된 결과에 대한 확신을 가질 수 있도록 설계 프로세스 전반에 걸쳐 매우 중요합니다.

일반적인 관행은 고전 이론, 실험 데이터, 게시 된 데이터, 유사한 구조의 성능 및 다른 사람이 수행 한 수치 계산에 대해 결과를 검증하는 것입니다. 때때로 소프트웨어 개발자가 제공 한 벤치 마크 또는 검증 예제가 이러한 목적으로 사용될 수 있지만 전체 범위의 문제를 포괄 할만큼 포괄적 인 경우는 거의 없습니다.

수치 해석을 시작하기 전에 분석가는 입력 데이터의 신뢰성, 소프트웨어 도구가 문제의 문제를 해결할 수 있는지 여부 및 결과를 검증하는 방법을 결정해야합니다. 검증 프로세스가 많은 실무자들에 의해 품질 보증 절차의 일부로 채택되었지만 비용이 많이 드는 실패가 여전히 발생했습니다 [1].

Validation

결과 검증의 필요성은 수치 분석의 사용 (남용)에서 일부 나쁜 업계 관행을 관찰함으로써 강화 될 수 있습니다. 수치 계산을 수행하기 위해 고용 된 일부 엔지니어 / 분석가는 계산 뒤에있는 기본 이론을 완전히 이해하지 못하거나 숨겨진 함정을 처리 할 수있는 실제 엔지니어링 경험이 충분하지 않을 수 있습니다.

일부 소프트웨어가 “CAD와 유사”해지고 많은 사람들이 작동하기 쉽다고 주장하기 때문에 엔지니어링 회사가 대학원 엔지니어 대신 초보를 고용하여 수치 모델링 및 분석을 수행하는 경향이 점차 증가하고 있습니다.

사용자는 복잡한 지오메트리 모델을 생성하고, 적절한 요소와 메시를 만들고, 각 하중 케이스에 대한 경계 조건 (접촉, 하중 및 고정)을 적용하고, 속성을 할당하고, 제출에 필요한 모든 플래그 / 스위치 / 버튼을 설정하는 데 상당한 노력을 기울일 것입니다.

분석이 실행됩니다. 자체 검사를위한 일부 품질 보증 절차는 전처리 단계에서 따를 수 있지만 계산이 완료되고 결과가 후 처리 될 때까지 많은 사용자는 출력이 어느 정도 정확하다고 쉽게 믿을 것입니다. 지오메트리 생성은 수치 모델링 프로세스의 일부일뿐입니다. 가장 어려운 문제 중 하나는 전체 설계 프로세스에서 불확실성을 다루는 것입니다. 재료 속성 및 로딩 순서와 같은 입력과 관련된 불확실성이 있습니다.

예를 들어 모델이 선형 또는 비선형 방식으로 동작하는지 여부와 같이 솔루션 유형의 적절성과 관련된 불확실성이 있습니다. 마지막으로 결과 해석과 관련된 불확실성이 있습니다. 수치 분석에서 결과를 검증하고 문제를 발견하는 데있어 분석가를위한 좋은 방법에 대한 간단한 지침은 없습니다. 그러나 다음 방법을 통해 점차적으로 달성 할 수 있습니다.

• 수치 적 방법 과정에 대한 좋은 이해 – 이것은 학부 및 / 또는 대학원 수준의 공식 교육을 통해 얻을 수 있으며 지속적인 전문성 개발의 일환으로 자습을 통해 더욱 향상 될 수 있습니다.
• 특정 유형의 문제에 대한 기본 이론과 해결책의 범위를 잘 이해합니다. 이 역시 위와 같은 교육을 통해 이루어질 수 있습니다.
• 실제 문제를 해결하는 데 공학적 판단을 사용하고 수치 분석을 수행 한 경험이 있습니다. 이는 숙련 된 엔지니어가 분석가를 적절하게 감독하는 환경에서 작업함으로써 얻을 수 있습니다.

품질 보증 시스템의 구현이 실행 가능한 솔루션으로 이어지는 엔지니어링 판단을 대체하는 것은 아니라는 점에 유의해야합니다. 복잡한 대규모 모델을 분석하기 전에 시뮬레이션 기술과 문제의 근본적인 동작을 완전히 이해하기 위해 간단한 테스트 모델을 사용하여 수치 “실험”을 수행해야하는 경우가 매우 많습니다.

경험에 따르면 때때로 테스트 모델 자체가 분석가가 최종 설계 솔루션에 도달 할 수있는 충분한 정보를 제공 할 수 있습니다. 해당 대형 복합 모델의 분석은 설계 기대치를 확인하는 것입니다. 다음 사례 연구는 결과 검증이 수행 된 방법과 신뢰 수준 및 불확실성이 해결된 방법을 보여줍니다.

Applications

일반적인 토목 공학 프로젝트에서 수치 분석은 구조 역학, 기하학 및 유체 역학의 세 가지 기본 분야 중 하나 또는 조합을 포함 할 수 있습니다. 문제의 성격은 토양-구조 상호 작용, 유체-구조 상호 작용 또는 토양-유체 상호 작용 중 하나로 분류 될 수 있습니다.

어떤 경우에는 세 가지 모두를 포함 할 수 있습니다. 잠재적 인 복잡성을 고려하여, 정확도를 잃지 않고 실제 목적을 위해 중요한 동작을 캡처하지 않고 문제를 단순화하기 위해 몇 가지 가정과 이상화가 이루어져야합니다. 이러한 문제를 해결할 수있는 범용 및 특수 수치 분석 소프트웨어가 있습니다. 두 가지 유형의 소프트웨어가 사례 연구에 사용되었습니다.

Case 1 – Deflection of a steel water tank

직경 약 90m의 대형 원형 강철 물 탱크는 처음 채울 때 큰 벽면이 휘어지면서 탱크의 장기적인 구조적 무결성에 대한 우려를 불러 일으켰습니다.

물의 높이는 전체 저장 용량에서 약 10m였습니다. 지붕 구조는 탱크 내부에있는 기둥으로 거의 전적으로지지되었습니다. 스트레이크(strakes)는 벽의 바닥 1/3이 더 두꺼운 고급 강판으로 구성되었습니다. 1 차 윈드 거더는 탱크 상단 주위에 용접되었고 2 차 윈드 거더는베이스 위 2/3에 위치했습니다. 하단 스트레이 크는 환형베이스 플레이트에 필렛 용접되었습니다. 내부 기둥의 기초를 제외한 전체 바닥은 용접 된 강판으로 덮여있었습니다.

이 탱크는 유능한 중간층 사암과 미사암 기반암 위에 압축된 채움물 위에 세워졌습니다. 일련의 축 대칭 유한 요소 분석 (FEA)을 수행하여 관찰된 처짐을 예측할 수 있는지 여부를 결정하고 매일 물을 채우고 비울 때 피로 파괴가 발생할 가능성으로 인해 벽 바닥의 응력 상태를 계산했습니다.

내부 기둥과 지붕 빔을 포함하는 탱크의 12 분의 1 섹터에 대한 3 차원 모델을 처음에 분석하여 벽이 얼마나 많은 지붕 자중을지지하고 축 대칭 가정의 타당성을 조사했는지 조사했습니다. 이 분석의 결과는 지붕 구조의 강성 기여도가 중요하지 않아 후속 축 대칭 모델에 포함되지 않았 음을 보여주었습니다.

그러나 지붕 자체 무게의 작은 부분이 벽에 적용됩니다. 축 대칭 모델은 모든 강철 섹션, 필렛 및 맞대기 용접 및 기초로 구성되었습니다 (그림 1). 그것들은 몇 개의 3 노드 삼각형 축 대칭 요소가있는 4 노드 비 호환 모드 사변형으로 이산화되었습니다.

용접 재료를 통해서만 하중 전달이 허용되도록 용접이 모델링되었습니다. 용접 연결부에 미세한 메시를 사용하여 응력 상태를 정확하게 포착했습니다. 롤러 지지대는 모델의 측면 및 하단 경계에 적용되었습니다. 다음과 같은 하중이 적용되었습니다 :

철골 구조물의 자중, 지붕 자중, 벽의 정수압, 수위에 따른 바닥의 균일 한 압력. 한 모델은 용접 또는베이스의 강판이 플라스틱 힌지를 형성하기 위해 항복되었다고 가정했습니다. 이 경우 벽 바닥에서 핀 연결이 모델링되었습니다.

Partial FE mesh of tank/foundation. Insert shows mesh and stress distribution at wall base
그림2 Partial FE mesh of tank/foundation. Insert shows mesh and stress distribution at wall base

벽 처짐은 그림 2에 나와 있습니다. 측정 범위와 계산 된 결과는 비교 목적으로 표시됩니다. 계산 된 벽 처짐을 검증하기 위해 두 벽 두께에 대한 Timoshenko 및 Woinowsky-Krieger [2]에 기반한 고전 이론도 그림에 표시되었습니다. 계산 된 편향은 이론적 계산에 의해 제한됨을 관찰 할 수 있습니다.

벽 두께의 변화로 인한 전이가 분석에서 포착되었습니다. 이것은 유한 요소 모델에 대한 확신을 제공했습니다. 윈드 거더와 구속 된베이스의 영향도 볼 수 있습니다. 윈드 거더 설치로 인해 초기 변형이 발생하여 공사가 끝날 때 벽 상단이 안쪽으로 당겨질 수 있습니다. 굽힘 동작이 발생한베이스 근처를 제외하고는 후프 동작이 벽 동작을 지배했습니다.

계산된 최대 처짐이 측정된 순서와 동일하더라도 최대 돌출이 발생한 높이는 예측되지 않았습니다. 실제로 조사 데이터는 몇 가지 가능한 시나리오를 제안했습니다.베이스에 플라스틱 힌지 형성 (그러나이 영역에서 계산 된 응력은 항복 강도를 초과하지 않았습니다). 지반 재료의 국부적 인 베어링 고장 (다시 현장에서 균열과 같은 명백한 지시 신호가 보이지 않음); 또는 탱크 건설이 끝날 때 내장 된 기하학적 결함이있었습니다. 사전 변형 된 탱크에서 역 분석을 수행하여 측정 된 처짐이 정수압 하에서 “회복”되었습니다. 그러나 계산된 응력은 수율을 훨씬 초과했습니다. 불행히도 탱크는 완성 후 첫 번째 충전 전에 즉시 조사되지 않았습니다.

Figure 2 Wall deflection of water tank
Figure 2 Wall deflection of water tank

탱크의 원래 디자인과 건설이 2000 년대 초에 수행되었다는 점은 흥미 롭습니다. 설계 계산에 관련 표준 [3]을 사용했습니다. 이 표준은 탱크 벽이 후프 동작만으로 작용한다고 가정하고이 구조의 경우가 아닌베이스의 제약 조건을 무시합니다. 벽 처짐의 크기는 기초 강성을 고려한 Rish [4]가 개발 한 고전 이론 [2] 또는 FEA와 같은 수치 분석에 의해 결정될 수 있습니다. 고급 강철을 사용하면 설계자는 강도에는 적합하지만 서비스 가능성에는 필요하지 않은 더 얇은 섹션을 선택해야합니다. 굽힘 강성은 큐브 두께에 의해 결정됩니다. 수중 부하에서 후속 벽 변형 프로파일은 제작 품질에 영향을받습니다. 이것은 설계 단계에서 추정하기 어려웠을 것입니다.

사례 2 – 배수로 배출

호주의 많은 댐 구조는 제한된 수 문학적 정보로 1950 년대와 60 년대에 설계 및 건설되었습니다. 이러한 기존 방수로 구조는 수정 된 가능한 최대 홍수 수준에 대처하기 위해 크기가 작습니다. 증가 된 홍수 조건 하에서 방수로 꼭대기에 대한 음압 생성과 같은 잠재적 인 문제가 발생할 수 있습니다. 이는 방수로 및 게이트 구조에 불안정성 또는 캐비테이션 손상을 유발할 수 있습니다. 역사적으로 스케일링 된 물리적 모델은 이러한 동작을 연구하기 위해 수력 학 실험실에서 구성되었지만 비용이 많이 들고 시간이 많이 걸리며 스케일링 효과와 관련된 많은 어려움이 있습니다. 오늘날 고성능 컴퓨터와보다 효율적인 전산 유체 역학 (CFD) 코드를 사용하여 수리적 구조의 동작을 합리적인 시간과 비용으로 수치 적으로 조사 할 수 있습니다. 이 분석 기법은 대도시 지역에 주요 상수원을 제공하는 가장 큰 콘크리트 중력 댐에 호주에서 처음으로 적용 되었기 때문에 검증을 수행 할 필요가있었습니다. 이것은 그림 3과 같이 조사 프로세스에 통합되었습니다. 순서도는 간단한 2D에서 상세한 3D 방수로 모델로 어떻게 발전했는지 보여줍니다.

Figure 3 Flowchart showing the validation process
Figure 3 Flowchart showing the validation process

미 육군 공병대 [5]에서 발표 한 광범위한 데이터가 있기 때문에 검증을 위해 ogee 방수로 프로필 (그림 4 참조)이 선택되었습니다. 계산 된 결과는 조사의 각 단계에서 검토되었습니다. 게시 된 데이터에서 크게 벗어나면 프로젝트가 중단됩니다. 이것은 프로젝트가 시작되기 전에 고객과 상호 합의되었습니다.

Figure 4 A view of the ogee spillway and Type 2 piers in the 3D CFD model
Figure 4 A view of the ogee spillway and Type 2 piers in the 3D CFD model

이러한 종류의 분석의 초기 어려움 중 하나는 개방 채널 중력 흐름 문제에서 자유 표면의 정확한 계산이었습니다. 자유 표면을 추적하는 데 적응 형 메싱 및 반복 방법을 사용하는 것은 일부 유한 체적 CFD 코드에서 사용되었지만 성공은 제한적이었습니다. 본 연구에 사용 된 코드는 SOLA-VOF 방법으로 Navier-Stokes 방정식을 해결합니다. 유체 운동의 과도 동작을 해결하기 위해 유한 차분 방법이 사용되었습니다. 유체의 부피 (VOF) 함수는 자유 표면 운동을 계산하는 데 사용됩니다 [6].

분석에 대한 자세한 내용은 [7]에 설명되어 있습니다. 계산 된 파고 압력 분포, 자유 표면 프로파일 및 정상 상태에서의 배출 속도는 검증 목적으로 사용되었습니다. 다른 상류 수두 (H) 아래의 배수로 꼭대기를 따라 압력 분포가 그림 5에 나와 있습니다. 일부 압력 진동은 코드가 일반 메시와 곡선 배수로 장애물 사이의 인터페이스에서 계산을 처리하는 방식에 기인 할 수 있습니다. 훨씬 더 미세한 메쉬는 이러한 불규칙성을 부드럽게 만들었습니다. 압력 분포에 대한 교각의 영향은 3D 모델에서 올바르게 예측되었습니다 (그림 6).

계산된 자유 표면 프로파일 (그림 7)도 게시 된 데이터와 잘 일치했습니다. Savage와 Johnson [8]은 분석 기법에 대한 신뢰도를 높이는 동일한 CFD 코드를 사용하여 유사한 유효성 검사를 수행했습니다. 문제의 배수로에 대한 후속 분석은 스케일링 된 물리적 모델 테스트에서 얻은 결과와 비교할 때 상당히 좋은 결과를 제공했습니다.

Figure 5 Comparison of crest pressure for various heads (2D model), Hd is the design head
Figure 5 Comparison of crest pressure for various heads (2D model), Hd is the design head
Figure 6 Comparison of crest pressure next to pier (3D model)
Figure 6 Comparison of crest pressure next to pier (3D model)
Figure 7 Upper nappe profile next to pier
Figure 7 Upper nappe profile next to pier

분석에서 배수로의 기하학적 구조와 물 속성이 잘 정의되었습니다. 물은 비압축성이며 고정 된 온도에서 일정한 특성을 가지고 있다고 가정했습니다. 실제로 좋은 품질의 콘크리트 표면 마감을 얻을 수 있기 때문에 배수로 경계는 매끄럽다 고 가정했습니다. 불확실성은 메쉬 밀도와 적절한 난류 모델의 선택이라는 두 가지 소스에서 비롯됩니다. 메쉬 크기는 메모리 양과 컴퓨터의 클럭 속도에 의해 제한됩니다.

높은 레이놀즈 수의 난류 흐름은 소용돌이와 소용돌이의 형성을 포착 할 수있는 매우 미세한 메시로 계산할 수 있지만 현재 메시 밀도는 검증 및 설계 목적에 필요한 변수를 예측하기에 충분히 미세했습니다. 조사 결과는 큰 와류, k-ε 및 RNG 모델과 같은 난류 모델의 선택에 의해 크게 영향을받지 않는 것으로 나타났습니다. 분명히 벽 거칠기와 난류 모델의 도입은 방전율을 감소시킬 것입니다. 그러나 다시 분석 결과는 사용 된 메시에 거의 영향을 미치지 않음을 보여줍니다. 향후 분석은 다른 메쉬 밀도로 인한 이산화 오류를 조사 할 것입니다.

사례 3 – 안벽 건설
주요 컨테이너 항구 시설은 설계 단계에서 최소한의 수치 분석을 수행하여 약 25 년 전에 건설되었습니다. 당시에는 이러한 분석 도구를 사용하는 것이 비용 효율적이지 않은 것으로 간주되었습니다. 다수의 컨테이너 크레인이 측면을 따라 이어지는 2km 길이의 안벽을 건설하기 위해 광범위한 준설 및 매립 작업이 수행되었습니다.

시설이 완공 된 이후 일련의 콘크리트 카운터 포트 유닛으로 구성된 안벽과 후방 크레인 빔은 크레인이 할 수 있도록 후방 빔에 대한 레벨 조정 작업이 수행 될 정도로 지속적으로 이동하고 있습니다. 정상적으로 작동합니다. 그러나 영향을받는 두 구조물의 움직임을 저지하기 위해보다 영구적 인 해결책을 모색했습니다. 토양-구조 상호 작용 및 시공 시뮬레이션을 처리 할 수있는 명시 적 유한 차이 분석을 사용하여 다양한 교정 옵션의 순위를 지정했습니다.

그라우트 기둥, 타이백 앵커 및 말뚝 지지대와 같은 다양한 제안 된 개선을 분석하기 전에, 토양 및 구조적 특성과 시공 과정의 선택이 적절하도록 계산 모델을 관찰에 대해 보정해야한다고 결정했습니다. 지질 및 지질 공학 정보는 현장 및 실험실 테스트 데이터를 포함하는 현장 조사 보고서에서 평가되었습니다. 시설의 범위를 고려할 때 현장에서 만나는 특정 토양 유형에 대해 상당한 분산 테스트 데이터가 예상됩니다. 수력 모래 충전재에 대한 표준 침투 테스트 (SPT) 블로우 횟수 (N) 및 콘 침투 테스트 (CPT) 저항 (qc)에 대한 몇 가지 일반적인 기록이 그림 8과 9에 나와 있습니다.

Figure 8 SPT ‘N’ profiles
Figure 8 SPT ‘N’ profiles
Figure 9 CPT profiles
Figure 9 CPT profiles

이 결과로부터 평균 해수면 위와 아래에있는 모래 채우기의 강도와 강성의 대비를 관찰 할 수 있습니다. 이 현상은 배치 방법에 기인한다고 제안되었다 [9]. 또한 기초 수준에서 진동 압축 된 모래의 특성에도 변동이있었습니다. 분석을 위해 선택된 토양 특성은 테스트 데이터, 인근 사이트의 경험 및 유사한 토양 조건에 대한 발표 된 데이터를 기반으로합니다. 그것들은 표 1에 요약되어 있습니다. 일반적으로 시설의 건설 순서는 다음과 같습니다.

  1. Removal of pockets of soft marine clay by dredging
  2. Dredging of sand to the required level
  3. Vibro-compaction of the sand on which the counterfort units were to be founded
  4. Placement of gravel for the quay wall foundation.
  5. Placement of concrete counterfort units weighing 360 tonne each
  6. Placement of hydraulic sand fill behind the units
  7. Surcharging the fill just behind the capping beam
  8. Construct capping beam and place more sand fill to the finished level
  9. Additional surcharge prior to the operation of container cranes.

Table 1 Soil properties used in the construction
simulation of the quay wall

Table 1 Soil properties used in the construction simulation of the quay wal
Table 1 Soil properties used in the construction simulation of the quay wal

2D 평면 변형 모델의 수치 시뮬레이션에서 구성 순서 (그림 10)와 하중은 다음 단계에 따라 단순화 / 이상적입니다.

  1. The starting condition of the seabed consisted of the vibrocompacted sand, gravel bed, native sand, clay and fissured clay at depth. The “in-situ” stresses were also switched on in this step.
  2. Placement of counterfort unit (using equivalent linear elastic beam elements) with a vertical force applied through the centre of gravity of the unit to represent the buoyant self-weight.
  3. Sequentially placing hydraulic sand fill behind the unit to the level prior to surcharging.
  4. Apply an equivalent trapezoidal pressure to represent the surcharge.
  5. Placement of capping beam and the sand fill to the required level.
  6. Apply additional surcharge.
  7. Application of repeated loads from the crane seaward and landward legs.
Figure 10 Construction sequence
Figure 10 Construction sequence

분석에서는 침수 된 물질과 평균 해수면 위에있는 물질을 나타 내기 위해 적절한 밀도를 사용했습니다. 안벽의 장기적인 움직임이 중요했기 때문에 배수 된 토양 매개 변수가 사용되었습니다. 토양은 분석에서 Mohr-Coulomb 실패 기준을 따르는 것으로 가정되었습니다. 단순한 탄성-완전 소성 응력-변형 거동이 가정되었습니다. 일련의 강체 다이어그램으로 표현 된 안벽 이동의 역사는 그림 11에 나와 있습니다. 벽의 상단과 바닥에서 계산 된 수직 및 수평 이동은 그림 12와 13에 표시됩니다. 수치는 모니터링 된 데이터와 해당 상한 및 하한 (해당 상자에 표시됨)입니다. 측정에서 산란의 양에도 불구하고 벽 건설에 대해 계산 된 움직임은 합리적으로 잘 비교되었습니다. 조사 데이터와 예측을 일치시키기 위해 분석에서 토양 속성을 변경하려는 시도가 없었습니다. 반복되는 크레인 하중의 래칫 효과를 관찰 할 수 있습니다. 불행히도 반복적 인 크레인 하중 하에서 벽 이동에 대한 기준이 없었기 때문에 이러한 예상 이동을 비교할 수 없었습니다. 문제의 복잡성과 고도로 가변적 인 토양 특성을 고려할 때 계산 된 결과는 매우 고무적입니다.

Figure 11 Wall deformations
Figure 11 Wall deformations

토양에서 플라스틱 구역의 발달도 분석에서 계산되었습니다. 벽의 발가락 아래의 토양이 여러 번 과도하게 압박을받는 것으로 밝혀졌습니다. 접촉 압력은 경사 하중으로 인한 베어링 고장에 대한 안전 지표 (FOS)를 결정하는 데 사용되었습니다. 지지력은 계산 방법에 의해 크게 영향을 받았다고보고되었습니다 [10]. 원래의 기초 디자인은 덴마크 코드 [11]를 기반으로했기 때문에이 경우 일관성을 위해 사용되었습니다. 편심의 함수로서 FOS의 발전과 수평 대 수직 추력 (H / V)의 비율이 각각 그림 14와 15에 나와 있습니다.

Figure 12 Wall top movements
Figure 12 Wall top movements
Figure 13 Wall base movements
Figure 13 Wall base movements
Figure 14 ‘FOS’ vs. eccentricity
Figure 14 ‘FOS’ vs. eccentricity
Figure 15 ‘FOS’ vs. H/V ratio
Figure 15 ‘FOS’ vs. H/V ratio

그림은 벽이 추가 요금과 반복적 인 적재 단계 동안 국부적 인 베어링 고장에 가까웠음을 보여줍니다. 크레인 하중 하에서 FOS의 명백한 증가는 벽에 대한 수직 하중이 증가하는 반면 유지된 토양의 수평 압력이 다소 일정하게 유지됨에 따라 편심이 감소했기 때문입니다.

끝 맺는 말
세 가지 매우 다른 실제 응용 프로그램의 유효성 검사 프로세스가 설명되었습니다. 각 사례의 주요 특징과 결과는 표 2에 요약되어 있습니다. 재료 및 하중 불확도 및 예상 결과가 강조 표시됩니다. 건설 품질은 구조의 성능에 상당한 영향을 미치는 것으로 나타났습니다.

이는 분석가가 프로젝트의 설계 단계에서 정량화하고 정확하게 분석하지 못할 수도 있습니다. 구조가 완료된 직후 모니터링의 중요성을 간과해서는 안됩니다. 이것은 미래의 역 분석을위한 유용한 자료가 될 것입니다. 수치 도구가 이러한 복잡한 문제를 분석 할 수 있다는 사실에도 불구하고 분석가는 어떤 매개 변수가 중요하거나 중요하지 않은지 식별 할 준비가되어 있어야합니다.

익숙하지 않은 문제를 분석 할 때 유효성 검사 프로세스를 점진적으로 수행해야합니다. 아마도 검증 방법을 찾는 핵심은 수치 분석 도구를 사용하지 않고 솔루션에 도달 할 수있는 다른 방법이 있는지 묻는 것입니다. 많은 경우 이러한 솔루션은 광범위한 문헌 검색 후에 존재합니다. 그러나 다른 경우에는 실험실 테스트와 현장 관찰이 유일한 대안이 될 것입니다.

자세한 내용은 원문을 참고하시기 바랍니다.

References
[1] Puri, S.P.S. (1998) “Avoiding Engineering Failures Caused by Computer-Related Errors”, J. Comp. in Civil Engineering, ASCE, 12(4), 170-172.
[2] Timoshenko, S.P. and Woinowsky-Krieger, S. (1959) Theory of Plates and Shells, 2nd edition, McGraw-Hill Kogakusha. p.580.
[3] BS2654 (1989) Manufacturing of vertical steel welded non-refrigerated storage tanks with butt-welded shells for the petroleum industry.
[4] Rish, R.F. (1977) “Design of Cylindrical Tanks on Elastic Foundations”, Civil Engineering Transactions, The Institution of Engineers, Australia, 192-195.
[5] US Army Corps of Engineers (1990) Hydraulic Design of Spillways, Engineer Manual No. 1110-2-1603.
[6] Hirt, C.W. and Nichols, B.D. (1981) “Volume of Fluid (VOF) Method for the Dynamics of Free Boundaries”, J. Comp. Phys. 39, 201- 225.
[7] Ho, D.K.H., Boyes, K.M and Donohoo, S.M. (2001) “Investigation of Spillway Behaviour under Increased Maximum Flood by Computational Fluid Dynamics Technique”, Proc. Conf. 14th Australasian Fluid Mechanics, Adelaide, December, 577-580.
[8] Savage, B.M. and Johnson, M.C. (2001) “Flow over Ogee Spillway: Physical and Numerical Model Case Study”, J. Hydraulic Engineering, ASCE, 127(8), 640-649.
[9] Lee, K.M., Shen, C.K., Leung, D.H.K. and Mitchell, J.K. (1999) “Effects of placement method on geotechnical behaviour of hydraulic fill sands” J. Geotech. and Geoenviron. Engineering, ASCE, 125(10), 832-846.
[10] Sieffert, J.G. and Bay-Gress, Ch. (2000) “Comparison of European bearing capacity calculation methods for shallow foundations”, Proceedings of the Institution of Civil Engineers, Geotechnical Engineering, 143, April, 65-74.
[11] DS 415 (1984) Code of Practice for Foundation Engineering. Table 2 Summary of findings for the three case studies

FLOW-3D 튜토리얼 V12

FLOW-3D 튜토리얼 V12

빠른 시작

이 튜토리얼 매뉴얼은 FLOW-3D 처음 사용하는 사용자에게 그래픽 사용자 인터페이스(GUI)의 주요 구성 요소를 쉽게 익히도록 하고, 다양한 시뮬레이션의 설정 및 실행 방법을 안내하기 위한 것입니다.

이 매뉴얼에 있는 실습과정은 FLOW-3D의 기본 사항을 다루기 위한 것입니다. 이 매뉴얼에서 제시하는 문제는 다양한 주제를 설명하고, 발생할 수 있는 많은 질문을 해결하기 위해 선정되었습니다. 이 매뉴얼의 실습과정은 FLOW-3D실행하는 컴퓨터에 앉아 사용하는 것이 가장 좋습니다.

CFD 사용 철학에 대한 간단한 섹션 다음에는 중요 파일과 시뮬레이션 파일을 실행하는 방법이 소개되어 있습니다. 이 소개 섹션 다음에는 모델 설정, 시뮬레이션 실행 및 포스트 프로세스, Simulation Manager 탐색 방법에 대한 설명이 있습니다. 이러한 각 단계에 대한 자세한 내용은 모델 설정, 컴퓨팅 결과 및 후처리 장에서 확인할 수 있습니다.

1.CFD 사용에 대한 철학

CFD (Computational Fluid Dynamics)는 유체 흐름(질량, 운동량 및 에너지 보존)에 대한 지배 방정식의 컴퓨터 솔루션입니다. 지정된 지배방정식은 이론 장에 설명된 Numerical방법을 사용하여 이산화되고 계산됩니다.

CFD 소프트웨어를 사용하는 것은 여러 면에서 실험을 설정하는 것과 유사합니다. 실제 상황을 시뮬레이션하기 위해 실험을 올바르게 설정하지 않으면, 그 결과는 실제 상황을 반영하지 않습니다. 같은 방법으로 수치 모델이 실제 상황을 정확하게 나타내지 않으면, 그 결과는 실제 상황을 반영하지 않습니다. 사용자는 어떤 것이 중요한지, 어떻게 표현해야 하는지를 결정해야 합니다. 시작하기 전에 다음과 같은 질문을 하는 것이 중요합니다.

  • CFD 계산에서 무엇을 알고 싶습니까?
  • 중요한 현상을 포착하기 위해 규모와 Mesh는 어떻게 설계되어야 하는가?
  • 실제 물리적 상황을 가장 잘 나타내는 경계 조건은 무엇입니까?
  • 어떤 종류의 유체를 사용해야합니까?
  • 이 문제에 어떤 유체 특성이 중요합니까?
  • 다른 어떤 물리적 현상이 중요합니까?
  • 초기 유체 상태는 어떻게 됩니까?
  • 어떤 단위 시스템을 사용해야합니까?

모델링 되는 문제가 실제 상황을 가능한 한 유사하게 나타내는지 확인하는 것이 중요합니다. 사용자는 복잡한 시뮬레이션 작업을 해결 가능한 부분으로 나누는 것이 좋습니다.

복잡한 물리 효과를 추가하기 전에, 간단하고 쉽게 이해할 수 있는 근사값으로 점차적으로 시작하여 프로세스 진행하십시오. 간단한 손 계산(베르누이 방정식, 에너지 균형, 파동
전파, 경계층 성장 등)은 물리 및 매개 변수를 선택하는데 도움이 되고, 결과와 비교할 수 있는 점검항목을 제공합니다.

CFD의 장단점을 이해하면 분석을 진행하는데 도움이 될 수 있습니다. CFD는 다음과 같은 경우 탁월한 분석 옵션입니다.

  • 기하 구조, 물리학 또는 필요한 상세 수준으로 인해 표준 엔지니어링 계산이 유용하지 않은 경우가 많습니다.
  • 실제 실험은 비용이 많이 소요됩니다.
  • 실험에서 수집할 수 있는 것보다 유체흐름에 대한 자세한 정보가 필요한 경우 유용합니다.
  • 위험하거나 적대적인 조건, 확장이 잘되지 않는 프로세스 등으로 인해 정확한 실험 측정을 하기가 어려운 경우
  • 복잡한 흐름 정보에 대한 커뮤니케이션

CFD는 다음과 같은 경우에 덜 효과적입니다.

  • 솔루션이 계산 리소스가 매우 많이 소요되거나, 도메인 크기를 줄이기 위한 가정 또는 해결되지 않은 물리적 현상을 설명하기 위한 반 임계 모델이 필요한 경우
  • CFD 시뮬레이션에 대한 입력이 되는 중요한 물리적 현상이 알려지지 않은 경우
  • 물리적 현상이 잘 이해되지 않거나 매우 복잡한 경우

CFD를 사용할 때 명심해야 할 몇 가지 중요한 참고 사항이 있습니다.

  • CFD는 규정된 초기 및 경계 조건에 따라 지정된 지배 방정식의 수치해석 솔루션입니다. 따라서 모델 설정, 즉 어떤 방정식을 풀어야 하는지, 재료 특성, 초기 조건 및 경계 조건이, 가능한 한 물리적 상황과 최대한 일치해야 합니다.
  • 방정식의 수치 해는 일반적으로 어떤 종류의 근사치를 필요로 합니다. 물리적 모델에 대한 가정과 해결방법을 검토한 후 사용하는 것이 좋습니다.
  • 디지털 컴퓨터는 숫자가 유한 정밀도로 이진수로 표시되는 방식으로 인해 반올림 오류가 발생합니다. 이는 문제를 악화시키기 때문에 매우 근소한 숫자의 차이를 계산해야 하는 상황을 피하십시오. 이러한 상황의 예는 시뮬레이션 도메인이 원점에서 멀리 떨어져 있을 때입니다.

 

2.중요한 파일

FLOW-3D 시뮬레이션과 관련된 많은 파일이 있습니다. 가장 중요한 것들이 아래에 설명되어 있습니다. 모든 prepin.* 파일의 명칭에서 prepin는 파일 형식을 의미하며, 별표시* 위치는 시뮬레이션 이름을 의미합니다. ( : prepin.example_simulation.)

  • ·prepin.*: 시뮬레이션용 입력 파일입니다. 시뮬레이션 설정을 설명하는 모든 입력 변수가 포함되어 있습니다.
  • ·prpgrf.*: 이것은 전 처리기 출력 파일입니다. 여기에는 계산된 초기 조건이 포함되며 시뮬레이션을 실행하기 전에 설정을 확인하는 데 사용될 수 있습니다.
  • ·flsgrf.*: 솔버 출력 파일입니다. 시뮬레이션의 최종 결과가 포함됩니다.
  • ·prperr.*, report.*, prpout.*: 이 파일들은 Preprocessor Diagnostic Files.
  • ·hd3err.*, hd3msg.*, hd3out.*: 이 파일들은 Solver Diagnostic Files.

모든 시뮬레이션 파일은 단일 폴더에 함께 유지하므로, 설명이 될 수 있는 시뮬레이션 이름을 사용하는 것이 좋습니다. 그러나 매우 긴 파일 이름은 운영 체제에 따라 문제가 될 수 있습니다.

노트

  • 시뮬레이션 이름이 inp(즉, 입력 파일이 있다면 prepin.inp) 출력 및 진단 파일은 모두 .dat이름을 갖습니다. 예: flsgrf.dat.
  • 모든 입력 파일은 네트워크 위치의 컴퓨터 대신 로컬 디렉토리에 저장하는 것이 좋습니다. 이것은 솔버가 더 빠르게 실행되고 GUI의 응답 속도가 빨라지며 실행중인 시뮬레이션을 방해하는 네트워크 문제 가능성을 제거합니다.

3.시뮬레이션 관리자

FLOW-3D 시뮬레이션 관리자의 탭은 주로 시뮬레이션을 실행할 수 있도록 시뮬레이션 환경을 구성하고 실행 시뮬레이션에 대한 상태 정보를 표시하는데 사용됩니다.

작업 공간 (Workspaces)

작업 공간(Workspaces)Simulation Manager의 필수 부분이며 파일을 FLOW-3D에서 처리하는 방식입니다. 기본적으로 시뮬레이션을 포함하고 구성하는 폴더입니다. 몇 가지 예를 들면 시뮬레이션과 또 다른 작업 공간인 검증 사례를 포함하도록 할 수 있습니다:

포트폴리오의 작업 공간

새로운 작업 공간 만들기

튜토리얼에서는 작성하려는 시뮬레이션을 포함할 작업 공간(Workspaces)을 작성하십시오.

1.File -> New workspace 이동

2.작업 공간 이름으로 Tutorial를 입력하십시오.

3.기본 위치는 현재 사용자의 홈 디렉토리에 있습니다. 다른 곳에서 찾을 수 있지만 기본 위치가 우리의 목적에 적합합니다.

4.하위 디렉토리를 사용하여 작업 공간 이름 만들기 확인란을 선택합니다. 이렇게 하면 파일 시스템에서 작업 공간에 대한 새로운 하위 디렉토리가 만들어져 시뮬레이션 파일을 훨씬 쉽게 구성할 수 있습니다.

새로운 작업 공간 만들기

5.확인을 눌러 새 작업 공간을 작성하십시오. 이제 포트폴리오에 표시됩니다.

새로운 작업 공간 만들기

작업 공간 닫기

포트폴리오를 정리하고 탐색하기 쉽도록 필요 없는 작업공간을 닫는 것이 편리합니다. 작업 공간을 닫으면 포트폴리오에서 해당 작업 공간만 제거됩니다. 그러나, 컴퓨터에서 작업 공간을 삭제하지는 않습니다.

작업 공간을 닫으려면

1.기존 작업 공간을 마우스 오른쪽 버튼으로 클릭하고 작업 Close Workspace 선택하십시오. 또는 포트폴리오에서 작업 공간을 선택 (왼쪽 클릭) 하고 Delete 키를 누를 수 있습니다.

2.작업 공간을 닫을 것인지 묻는 메세지가 표시됩니다. 예를 선택하십시오.

3.포트폴리오는 더 이상 닫힌 작업 공간을 포함하지 않습니다.

기존 작업 공간 열기

오래된 작업 공간을 열어야 할 때가 있을 것입니다. 예를 들어, 새 프로젝트에 유사한 시뮬레이션을 작성하기 전에 기존 시뮬레이션의 설정을 검토할 수 있습니다. 기존 작업 공간을 열려면

1.File -> Open Workspace를 선택하십시오

2.작업 공간 파일이 있는 디렉토리를 찾으십시오. Tutorial.FLOW-3D_Workspace.

작업 공간 열기

3.작업 공간을 로드 하려면 OK누르십시오.

작업 공간에서 시뮬레이션 작업