THE APPLICATION OF MODIFIED PHYSICS-INFORMED NEURAL NETWORKS IN RAYLEIGH-TAYLOR INSTABILITY

Qiu Rundi; Wang Jingzhu; Huang Renfang; Du Tezhuan; Wang Yiwei; Huang Chenguang

doi:10.6052/0459-1879-22-253

Volume 54 Issue 8

Aug. 2022

Turn off MathJax

Article Contents

Abstract

References

Chinese Journal of Theoretical and Applied Mechanics > 2022 > 54(8): 2224-2234. > DOI: 10.6052/0459-1879-22-253 CSTR: 32045.14.0459-1879-22-253

Qiu Rundi, Wang Jingzhu, Huang Renfang, Du Tezhuan, Wang Yiwei, Huang Chenguang. The application of modified physics-informed neural networks in Rayleigh-Taylor instability. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(8): 2224-2234. DOI: 10.6052/0459-1879-22-253

Citation:

PDF (2991 KB)

THE APPLICATION OF MODIFIED PHYSICS-INFORMED NEURAL NETWORKS IN RAYLEIGH-TAYLOR INSTABILITY

Qiu Rundi^{*, †},
Wang Jingzhu^{*, **},
Huang Renfang^*,
Du Tezhuan^{*, **},
Wang Yiwei^{*, †, **,},
Huang Chenguang^{*, **, ††}

*.
Key Laboratory Mechanics in Fluid Solid Coupling Systems, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, China
†.
School of Future Technology, University of Chinese Academy of Sciences, Beijing 100049, China
**.
School of Engineering Science, University of Chinese Academy of Sciences, Beijing 100049, China
††.
Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China

Received Date: June 05, 2022
Accepted Date: July 14, 2022
Available Online: July 17, 2022

Graphical Abstract

Abstract

Abstract

Physics-informed neural networks for phase-field method (PF-PINNs) are successfully applied to the modeling of two-phase flow, and it provides a brand-new way for the high-accuracy direct numerical simulation of two-phase flow. As an emerging interface capturing method, the introduction of the phase-field method in two phase flow ensures mass conservation near the interface and significantly enhances the interface capturing accuracy. However, the high-order derivate introduced by the phase-field method decreases the efficiency of network training. To enhance the efficiency of the training process, this paper regards the chemical energy as an auxiliary parameter and one of the outputs of the proposed neural network and revises the loss functions of physics constrain in the PF-PINNs framework based on the deep mixed residual method (MIM), which transforms the relationship between the auxiliary parameter and the phase-field variable from hard constrain to soft constrain. The proposed improvements decrease the size of the computational graph generated in automatic differentiation significantly and the computational cost of calculating high-order derivates in automatic differentiation is reduced. Meanwhile, the Rayleigh-Taylor (RT) instability is tested to assess the modeling ability of the proposed PF-PINNs when the Reynolds number is high and an enormous amount of calculation is needed. Compared with the spectral element-based phase-field method qualitatively and quantitatively, modified PF-PINNs can capture the strong non-linear evolution process of the interface, and the accuracy of modified PF-PINNs reaches the accuracy of traditional numerical solver. The result of the proposed neural network fits the characteristic of RT instability well. Compared the modified PF-PINNs with the original PF-PINNs, the results indicate that the deep mixed residual method can reduce the training time of the original PF-PINNs notably. The proposed method in this paper is a valuable reference for improving the training speed of the neural network and gives a new insight into exploring the intelligent modeling method with high accuracy.
- Rayleigh-Taylor instability,
- deep mixed residual method,
- physics-informed neural networks,
- two-phase flow

FullText(HTML)

References (33)

References

[1]	Magnaudet J, Mercier MJ. Particles, drops, and bubbles moving across sharp interfaces and stratified layers. Annual Review of Fluid Mechanics, 2020, 52(1): 61-91 doi: 10.1146/annurev-fluid-010719-060139
[2]	Ruspini LC, Marcel CP, Clausse A. Two-phase flow instabilities: A review. International Journal of Heat and Mass Transfer, 2014, 71: 521-548 doi: 10.1016/j.ijheatmasstransfer.2013.12.047
[3]	Sui Y, Ding H, Spelt PDM. Numerical simulations of flows with moving contact lines. Annual Review of Fluid Mechanics, 2014, 46(1): 97-119 doi: 10.1146/annurev-fluid-010313-141338
[4]	Huang B, Qiu SC, Li XB, et al. A review of transient flow structure and unsteady mechanism of cavitating flow. Journal of Hydrodynamics, 2019, 31(3): 429-444 doi: 10.1007/s42241-019-0050-0
[5]	郭子漪, 李凯, 康琦等. 界面张力梯度驱动对流向湍流转捩的研究. 力学进展, 2021, 51(1): 1-28 doi: 10.6052/1000-0992-20-022 Guo Ziyi, Li Kai, Kang Qi et. al. Study on bifurcation to chaos of surface tension gradient driven flow. Advances in Mechanics, 2021, 51(1): 1-28 (in Chinese) doi: 10.6052/1000-0992-20-022
[6]	李康, 刘娜, 何志伟等. 一种基于双界面函数的界面捕捉方法. 力学学报, 2017, 49(6): 1290-1300 doi: 10.6052/0459-1879-17-210 Li Kang, Liu Na, He Zhiwei et. al. A new interface capturing method based on double interface functions. Chinese Journal of Theoretical and Applied Mechanics, 2017, 49(6): 1290-1300 (in Chinese) doi: 10.6052/0459-1879-17-210
[7]	Hirt CW, Nichols BD. Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics, 1981, 39(1): 201-225 doi: 10.1016/0021-9991(81)90145-5
[8]	Sethian JA, Smereka P. Level set methods for fluid interfaces. Annual Review of Fluid Mechanics, 2003, 35(1): 341-372 doi: 10.1146/annurev.fluid.35.101101.161105
[9]	Sussman M, Smereka P, Osher S. A level set approach for computing solutions to incompressible two-phase flow. Journal of Computational Physics, 1994, 114(1): 146-159 doi: 10.1006/jcph.1994.1155
[10]	Anderson DM, Mcfadden GB, Wheeler AA. Diffuse-interface methods in fluid mechanics. Annual Review of Fluid Mechanics, 1998, 30(1): 139-165 doi: 10.1146/annurev.fluid.30.1.139
[11]	Gurtin ME, Polignone D, ViÑals J. Two-phase binary fluids and immiscible fluids described by an order parameter. Mathematical Models and Methods in Applied Sciences, 1996, 6: 815-831 doi: 10.1142/S0218202596000341
[12]	Qiao L, Zeng Z, Xie HQ, et al. Modeling thermocapillary migration of interfacial droplets by a hybrid lattice Boltzmann finite difference scheme. Applied Thermal Engineering, 2018, 131: 910-919 doi: 10.1016/j.applthermaleng.2017.12.034
[13]	Yue PT. Thermodynamically consistent phase-field modelling of contact angle hysteresis. Journal of Fluid Mechanics, 2020, 899: A15 doi: 10.1017/jfm.2020.465
[14]	Taylor GI. The instability of liquid surfaces when accelerated in a direction perpendicular to their planes. I. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 1950, 201(1065): 192-196
[15]	Rayleigh R. Investigation of the character of the equilibrium of an incompressible heavy fluid of variable density. Proceedings of the London Mathematical Society, 1882, 1(1): 170-177
[16]	Zhao Z, Wang P, Liu N, et al. Analytical model of nonlinear evolution of single-mode Rayleigh–Taylor instability in cylindrical geometry. Journal of Fluid Mechanics, 2020, 900: A24
[17]	Zhang TW, Wu J, Lin XJ. An interface-compressed diffuse interface method and its application for multiphase flows. Physics of Fluids, 2019, 31(12): 122102 doi: 10.1063/1.5116035
[18]	赵暾, 周宇, 程艳青等. 基于内嵌物理机理神经网络的热传导方程的正问题及逆问题求解. 空气动力学学报, 2021, 39(5): 19-26 doi: 10.7638/kqdlxxb-2020.0176 Zhao Tun, Zhou Yu, Cheng Yanqing et. al. Solving forward and inverse problems of the heat conduction equation using physics-informed neural networks. Acta Aerodynamica Sinica, 2021, 39(5): 19-26(in Chinese)) doi: 10.7638/kqdlxxb-2020.0176
[19]	陆志彬, 瞿景辉, 刘桦等. 基于物理信息神经网络的传热过程物理场代理模型的构建. 化工学报, 2021, 72(3): 1496-1503 Lu Zhibin, Qu Jinghui, Liu Hua et. al. Surrogate modeling for physical fields of heat transfer processes based on physics-informed neural network. CIESC Journal, 2021, 72(3): 1496-1503(in Chinese))
[20]	金晓威, 赖马树金, 李惠. 物理增强的流场深度学习建模与模拟方法. 力学学报, 2021, 53(10): 2616-2629 doi: 10.6052/0459-1879-21-373 Jin Xiaowei, Laima Shujin, Li Hui. Physics-enhanced deep learning methods for modelling and simulating flow fields. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(10): 2616-2629(in Chinese)) doi: 10.6052/0459-1879-21-373
[21]	Karniadakis GE, Kevrekidis IG, Lu L, et al. Physics-informed machine learning. Nature Reviews Physics, 2021, 3(6): 422-440 doi: 10.1038/s42254-021-00314-5
[22]	Raissi M, Perdikaris P, Karniadakis GE. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational physics, 2019, 378: 686-707 doi: 10.1016/j.jcp.2018.10.045
[23]	Raissi M, Yazdani A, Karniadakis GE. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science, 2020, 367(6481): 1026-1030 doi: 10.1126/science.aaw4741
[24]	Cai SZ, Wang ZC, Fuest F, et al. Flow over an espresso cup: inferring 3-D velocity and pressure fields from tomographic background oriented Schlieren via physics-informed neural networks. Journal of Fluid Mechanics, 2021, 915: A102 doi: 10.1017/jfm.2021.135
[25]	Buhendwa AB, Adami S, Adams NA. Inferring incompressible two-phase flow fields from the interface motion using physics-informed neural networks. Machine Learning with Applications, 2021, 4: 100029 doi: 10.1016/j.mlwa.2021.100029
[26]	Qiu RD, Huang RF, Xiao Y, et al. Physics-informed neural networks for phase-field method in two-phase flow. Physics of Fluids, 2022, 34(5): 052109 doi: 10.1063/5.0091063
[27]	Lyu LY, Zhang Z, Chen MX, et al. MIM: A deep mixed residual method for solving high-order partial differential equations. Journal of Computational Physics, 2022, 452: 110930 doi: 10.1016/j.jcp.2021.110930
[28]	Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv, 2014: 1412.6980
[29]	Glorot X, Bengio Y. Understanding the difficulty of training deep feedforward neural networks//Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 2010: 249-256
[30]	Wight CL, Zhao J. Solving Allen-Cahn and Cahn-Hilliard equations using the adaptive physics informed neural networks. Communications in Computational Physics, 2021, 29(3): 930-954 doi: 10.4208/cicp.OA-2020-0086
[31]	Tryggvason G. Numerical simulations of the Rayleigh-Taylor instability. Journal of Computational Physics, 1988, 75(2): 253-282 doi: 10.1016/0021-9991(88)90112-X
[32]	Huang ZY, Lin G, Ardekani AM. Consistent, essentially conservative and balanced-force phase-field method to model incompressible two-phase flows. Journal of Computational Physics, 2020, 406: 109192 doi: 10.1016/j.jcp.2019.109192
[33]	Xiao Y, Zeng Z, Zhang L, et al. A spectral element-based phase field method for incompressible two-phase flows. Physics of Fluids, 2022, 34(2): 022114 doi: 10.1063/5.0077372

[1]	Wang Changli, Zhao Yanying. THEORY AND EXPERIMENT OF EQUAL-PEAK OPTIMIZATION OF TIME DELAY COUPLED PENDULUM TUNED MASS DAMPER VIBRATING SYSTEM[J]. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(4): 954-971. DOI: 10.6052/0459-1879-23-026
[2]	Zeng Jianjian, Zheng Yuanguang. STABILIZING UNSTABLE PERIODIC TRAJECTORIES OF CHAOTIC SYSTEMS WITH TIME-VARYING SWITCHING DELAYED FEEDBACK CONTROL[J]. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(12): 3477-3485. DOI: 10.6052/0459-1879-22-361
[3]	Dai Han, Zhao Yanying. EQUAL-PEAK OPTIMIZATION OF DYNAMIC VIBRATION ABSORBER WITH NEGATIVE STIFFNESS AND DELAY FEEDBACK CONTROL[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(6): 1720-1732. DOI: 10.6052/0459-1879-21-074
[4]	Jin Yanfei, Wang Heqiang. DYNAMICAL ANALYSIS OF A PERIODIC POTENTIAL DRIVEN BY ADDITIVE AND MULTIPLICATIVE TRICHOTOMOUS NOISES[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(3): 865-873. DOI: 10.6052/0459-1879-20-199
[5]	Wang Zaihua, Hu Haiyan. STABILITY OF A FORCE CONTROL SYSTEM WITH SAMPLED-DATA FEEDBACK[J]. Chinese Journal of Theoretical and Applied Mechanics, 2016, 48(6): 1372-1381. DOI: 10.6052/0459-1879-16-102
[6]	Yuanguang Zheng Chengdai Huang Zaihua Wang. Delay effect on the relaxation oscillations of a van der pol oscillator with delayed feedback[J]. Chinese Journal of Theoretical and Applied Mechanics, 2012, 44(1): 148-157. DOI: 10.6052/0459-1879-2012-1-lxxb2011-244
[7]	Yanying Zhao Jian Xu. Using the delayed feedback to control the vibration of the auto-parametric dynamical system[J]. Chinese Journal of Theoretical and Applied Mechanics, 2011, 43(5): 894-904. DOI: 10.6052/0459-1879-2011-5-lxxb2010-652
[8]	ew Features of Positive Time-Delayed Feedbacks in Vibration Control[J]. Chinese Journal of Theoretical and Applied Mechanics, 2010, 42(5): 933-942. DOI: 10.6052/0459-1879-2010-5-lxxb2009-313
[9]	ON CO-DIMENSION 2 BIFURCATION SYSTEM EXCITED PARAMETRICALLY BY WHITE NOISES[J]. Chinese Journal of Theoretical and Applied Mechanics, 1997, 29(5): 563-572. DOI: 10.6052/0459-1879-1997-5-1995-267
[10]	ON THE HOPF BIFURCATION SYSTEM IN THE PRESENCE OF PARAMETRIC REAL NOISES[J]. Chinese Journal of Theoretical and Applied Mechanics, 1997, 29(2): 158-166. DOI: 10.6052/0459-1879-1997-2-1995-211

Cited By

Cited by

Periodical cited type(10)

1.	王美琪，曾思恒，李源，刘鹏飞. 二自由度磁浮列车悬浮系统时滞控制研究. 西南交通大学学报. 2024(04): 812-822 .
2.	周碧柳，靳艳飞. 高斯色噪声和谐波激励共同作用下耦合SD振子的混沌研究. 力学学报. 2022(07): 2030-2040 . 本站查看
3.	郭筱瑛，周英姿，王利华，王诗心. 具有随机黏性阻尼的分数维线性振荡器中的随机多共振. 电工技术学报. 2021(03): 532-541 .
4.	靳艳飞，王贺强. 加性和乘性三值噪声激励下周期势系统的动力学分析. 力学学报. 2021(03): 865-873 . 本站查看
5.	朱建渠，金炜东，郭锋. 乘性噪声作用下延迟分数阶系统中的随机共振. 西南交通大学学报. 2021(02): 363-370 .
6.	张婉洁，牛江川，申永军，杨绍普，王丽. 一类阻尼控制半主动隔振系统的解析研究. 力学学报. 2020(06): 1743-1754 . 本站查看
7.	吴晗，曾晓辉，史禾慕. 考虑间隙反馈控制时滞的磁浮车辆稳定性研究. 力学学报. 2019(02): 550-557 . 本站查看
8.	芮珍梅，陈建兵. 加性非平稳激励下结构速度响应的FPK方程降维. 力学学报. 2019(03): 922-931 . 本站查看
9.	蓝昱群，关利南，古华光. 负反馈诱发神经电振荡的反常现象的复杂动力学. 力学学报. 2019(04): 1122-1133 . 本站查看
10.	邢子康，申永军，李向红. 接地式三要素型动力吸振器性能分析. 力学学报. 2019(05): 1466-1475 . 本站查看

Other cited types(2)

Get Citation

PDF

XML

Article Metrics

Article views (1383) PDF downloads (229) Cited by(12)

THE APPLICATION OF MODIFIED PHYSICS-INFORMED NEURAL NETWORKS IN RAYLEIGH-TAYLOR INSTABILITY

Abstract

References

Related Articles

Cited by

Periodical cited type(10)

Other cited types(2)

Catalog

Article Metrics

Related

Download Center

Author Center

THE APPLICATION OF MODIFIED PHYSICS-INFORMED NEURAL NETWORKS IN RAYLEIGH-TAYLOR INSTABILITY

Abstract

References

Related Articles

Cited by

Periodical cited type(10)

Other cited types(2)

Catalog

Article Metrics

Related

Download Center

Author Center

Export File

Citation

Format

Content