Deep energy method in topology optimization applications. (English) Zbl 1518.74067
Summary: This paper explores the possibilities of applying physics-informed neural networks (PINNs) in topology optimization (TO) by introducing a fully self-supervised TO framework based on PINNs. This framework solves the forward elasticity problem by the deep energy method (DEM). Instead of training a separate neural network to update the density distribution, we leverage the fact that the compliance minimization problem is self-adjoint to express the element sensitivity directly in terms of the displacement field from the DEM model. Thus, no additional neural network is needed for the inverse problem. The method of moving asymptotes is used as the optimizer for updating density distribution. The implementation of Neumann, Dirichlet, and periodic boundary conditions is described in the context of the DEM model. Three numerical examples are presented to demonstrate framework capabilities: (i) compliance minimization in 2D under different geometries and loading, (ii) compliance minimization in 3D, and (iii) maximization of homogenized shear modulus to design 2D metamaterial unit cells. The results show that the optimized designs from the DEM-based framework are very comparable to those generated by the finite element method and shed light on a new way of integrating PINN-based simulation methods into classical computational mechanics problems.
MSC:
| 74P15 | Topological methods for optimization problems in solid mechanics |
| 74P05 | Compliance or weight optimization in solid mechanics |
| 74S99 | Numerical and other methods in solid mechanics |
| 68T05 | Learning and adaptive systems in artificial intelligence |
Keywords:
physics-informed neural network; compliance minimization; method of moving asymptotes; homogenized shear modulus maximizationReferences:
| [1] | Chen, C-T; Gu, GX, Machine learning for composite materials, MRS Commun., 9, 2, 556-566 (2019) · doi:10.1557/mrc.2019.32 |
| [2] | Yang, C., Kim, Y., Ryu, S., Gu, G.X.: Prediction of composite microstructure stress-strain curves using convolutional neural networks. Mater. Des. 189:108509 (2020) |
| [3] | Luo, L.; Zhang, B.; Zhang, G.; Li, X.; Fang, X.; Li, W.; Zhang, Z., Rapid prediction and inverse design of distortion behaviors of composite materials using artificial neural networks, Polym. Adv. Technol., 32, 3, 1049-1060 (2021) · doi:10.1002/pat.5152 |
| [4] | He, J., Kushwaha, S., Abueidda, D., Jasiuk, I.: Exploring the structure-property relations of thin-walled, 2d extruded lattices using neural networks. Comput Struct. (2022), in press. doi:10.1016/j.compstruc.2022.106940 |
| [5] | Chen, G.; Li, T.; Chen, Q.; Ren, S.; Wang, C.; Li, S., Application of deep learning neural network to identify collision load conditions based on permanent plastic deformation of shell structures, Comput. Mech., 64, 2, 435-449 (2019) · Zbl 1468.74082 · doi:10.1007/s00466-019-01706-2 |
| [6] | Stoffel, M.; Bamer, F.; Markert, B., Neural network based constitutive modeling of nonlinear viscoplastic structural response, Mech. Res. Commun., 95, 85-88 (2019) · Zbl 1425.74504 · doi:10.1016/j.mechrescom.2019.01.004 |
| [7] | Maysam B, Gorji, Mojtaba, Mozaffar, Julian N, Heidenreich, Jian, Cao, and Dirk, Mohr: On the potential of recurrent neural networks for modeling path dependent plasticity. J. Mech. Physics Solids, 143: 103972 (2020) |
| [8] | Diab W, Abueidda, Seid, Koric, Nahil A, Sobh, and Huseyin, Sehitoglu. Deep learning for plasticity and thermo-viscoplasticity. Int. J. Plasticity, 136:102852 (2021) |
| [9] | Chen, G., Recurrent neural networks (rnns) learn the constitutive law of viscoelasticity, Comput. Mech., 67, 3, 1009-1019 (2021) · Zbl 1494.74011 · doi:10.1007/s00466-021-01981-y |
| [10] | Yang, H.; Xiang, Q.; Tang, S.; Guo, X., Learning material law from displacement fields by artificial neural network, Theor. Appl. Mech. Lett., 10, 3, 202-206 (2020) · doi:10.1016/j.taml.2020.01.038 |
| [11] | Flaschel, M.; Kumar, S.; De Lorenzis, L., Unsupervised discovery of interpretable hyperelastic constitutive laws, Comput. Methods Appl. Mech. Eng., 381 (2021) · Zbl 1506.74051 · doi:10.1016/j.cma.2021.113852 |
| [12] | Khemraj, Shukla, Mengjia, Xu, Nathaniel, Trask, and George E, Karniadakis: Scalable algorithms for physics-informed neural and graph networks. Data-Centric Engineering, 3 (2022) |
| [13] | Haghighat, E.; Raissi, M.; Moure, A.; Gomez, H.; Juanes, R., A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics, Comput. Methods Appl. Mech. Eng., 379 (2021) · Zbl 1506.74476 · doi:10.1016/j.cma.2021.113741 |
| [14] | Shengze, Cai, Zhiping, Mao, Zhicheng, Wang, Minglang, Yin, and George Em, Karniadakis. Physics-informed neural networks (pinns) for fluid mechanics: A review. Acta Mech. Sinica, 1-12 (2022) |
| [15] | Henkes, A.; Wessels, H.; Mahnken, R., Physics informed neural networks for continuum micromechanics, Comput. Methods Appl. Mech. Eng., 393 (2022) · Zbl 1507.74478 · doi:10.1016/j.cma.2022.114790 |
| [16] | Raissi, M., Deep hidden physics models: Deep learning of nonlinear partial differential equations, J. Machine Learning Res, 19, 1, 932-955 (2018) · Zbl 1439.68021 |
| [17] | Diab, W, Abueidda, Qiyue, Lu, and Seid, Koric: Meshless physics-informed deep learning method for three-dimensional solid mechanics. Int. J. Numerical Methods Eng. 122(23):7182-7201 (2021) · Zbl 07863835 |
| [18] | Hongwei, Guo, Xiaoying, Zhuang, and Timon, Rabczuk: A deep collocation method for the bending analysis of Kirchhoff plate. arXiv preprint arXiv:2102.02617 (2021) |
| [19] | Yan, CA; Vescovini, R.; Dozio, L., A framework based on physics-informed neural networks and extreme learning for the analysis of composite structures, Comp. Struc., 265 (2022) · doi:10.1016/j.compstruc.2022.106761 |
| [20] | Sina Amini, Niaki, Ehsan, Haghighat, Trevor, Campbell, Anoush, Poursartip, and Reza, Vaziri: Physics-informed neural network for modelling the thermochemical curing process of composite-tool systems during manufacture. Computer Methods Appl. Mech. Eng. 384:113959 (2021) · Zbl 1506.74123 |
| [21] | Wessels, H.; Weißenfels, C.; Wriggers, P., The neural particle method-an updated Lagrangian physics informed neural network for computational fluid dynamics, Comput. Methods Appl. Mech. Eng., 368 (2020) · Zbl 1506.76136 · doi:10.1016/j.cma.2020.113127 |
| [22] | Bing, Yu., et al.: The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems. Com. Math. Statistics 6(1), 1-12 (2018) · Zbl 1392.35306 |
| [23] | Yulei, Liao and Pingbing, Ming: Deep nitsche method: Deep Ritz method with essential boundary conditions. arXiv preprint arXiv:1912.01309 (2019) |
| [24] | Larry J Segerlind. Applied Finite Element Analysis. (1984) · Zbl 0404.73064 |
| [25] | Junuthula Narasimha, Reddy: An Introduction to Nonlinear Finite Element Analysis, Second Edition: With Applications to Heat Transfer, Fluid Mechanics, and Solid Mechanics. OUP Oxford (2014) · Zbl 1305.65002 |
| [26] | Esteban, Samaniego, Cosmin, Anitescu, Somdatta, Goswami, Vien Minh, Nguyen-Thanh, Hongwei, Guo, Khader, Hamdia, X Zhuang, and T Rabczuk: An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applications. Computer Methods in Applied Mechanics and Engineering, 362:112790 (2020) · Zbl 1439.74466 |
| [27] | Vien Minh, Nguyen-Thanh, Xiaoying, Zhuang, and Timon Rabczuk: A deep energy method for finite deformation hyperelasticity. European J. Mech.-A/Solids, 80:103874 (2020) · Zbl 1472.74213 |
| [28] | Jan N, Fuhg, and Nikolaos Bouklas: The mixed deep energy method for resolving concentration features in finite strain hyperelasticity. J. Comp. Phy. 451:110839 (2022) · Zbl 07517152 |
| [29] | Diab W, Abueidda, Seid, Koric, Erman, Guleryuz, and Nahil A, Sobh: Enhanced physics-informed neural networks for hyperelasticity. arXiv preprint arXiv:2205.14148 (2022a) |
| [30] | Shahed, Rezaei, Ali, Harandi, Ahmad, Moeineddin, Bai-Xiang, Xu, and Stefanie Reese: A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method. arXiv preprint arXiv:2206.13103 (2022) · Zbl 1507.74068 |
| [31] | Hunter T, Kollmann, Diab W, Abueidda, Seid, Koric, Erman, Guleryuz, and Nahil A, Sobh: Deep learning for topology optimization of 2d metamaterials. Materials & Design, 196:109098 (2020) |
| [32] | Diab W, Abueidda, Seid, Koric, and Nahil A, Sobh: Topology optimization of 2d structures with nonlinearities using deep learning. Computers & Struct. 237:106283 (2020) |
| [33] | Sosnovik, I.; Oseledets, I., Neural networks for topology optimization, Russ. J. Numer. Anal. Math. Model., 34, 4, 215-223 (2019) · Zbl 1420.68178 · doi:10.1515/rnam-2019-0018 |
| [34] | Saurabh, Banga, Harsh, Gehani, Sanket, Bhilare, Sagar, Patel, and Levent, Kara: 3d topology optimization using convolutional neural networks. arXiv preprint arXiv:1808.07440 (2018) |
| [35] | Zeyu Zhang, Yu; Li, WZ; Chen, X.; Yao, W.; Zhao, Y., Tonr: An exploration for a novel way combining neural network with topology optimization, Comput. Methods Appl. Mech. Eng., 386 (2021) · Zbl 1507.74346 · doi:10.1016/j.cma.2021.114083 |
| [36] | Stephan, Hoyer, Jascha, Sohl-Dickstein, and Sam Greydanus: Neural reparameterization improves structural optimization. arXiv preprint arXiv:1909.04240 (2019) |
| [37] | Chandrasekhar, A.; Suresh, K., Tounn: topology optimization using neural networks, Struct. Multidiscip. Optim., 63, 3, 1135-1149 (2021) · doi:10.1007/s00158-020-02748-4 |
| [38] | Zehnder, J.; Li, Y.; Coros, S.; Thomaszewski, B., Ntopo: Mesh-free topology optimization using implicit neural representations, Adv. Neural. Inf. Process. Syst., 34, 10368-10381 (2021) |
| [39] | Pattanayak, S.: John S Pattanayak, and Suresh John. Pro Deep Learning with Tensorflow. Springer (2017) |
| [40] | Ciyou, Zhu, Richard H, Byrd, Peihuang, Lu, and Jorge, Nocedal. Algorithm 778: L-bfgs-b: Fortran subroutines for large-scale bound-constrained optimization. ACM Transactions on Mathematical Software (TOMS), 23 (4):550-560 (1997) · Zbl 0912.65057 |
| [41] | Joshua M. Long: Random Fourier Features Pytorch. GitHub. Note: https://github.com/jmclong/random-fourier-features-pytorch (2021) |
| [42] | Wang, S.; Wang, H.; Perdikaris, P., On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale pdes with physics-informed neural networks, Comput. Methods Appl. Mech. Eng., 384 (2021) · Zbl 1506.35130 · doi:10.1016/j.cma.2021.113938 |
| [43] | Chadha, C.; Abueidda, D.; Koric, S.; Guleryuz, E.; Jasiuk, I., Optimizing hyperparameters and architecture of deep energy method. (2022) · doi:10.20944/preprints202206.0414.v1 |
| [44] | Diab W, Abueidda, Seid, Koric, Rashid Abu, Al-Rub, Corey M, Parrott, Kai A, James, and Nahil A, Sobh: A deep learning energy method for hyperelasticity and viscoelasticity. European J. Mech.-A/Solids, 95:104639 (2022b) · Zbl 1490.74120 |
| [45] | Vien Minh, Nguyen-Thanh, Cosmin, Anitescu, Naif, Alajlan, Timon, Rabczuk, and Xiaoying, Zhuang: Parametric deep energy approach for elasticity accounting for strain gradient effects. Comp. Methods Appl. Mech. Eng. 386:114096 (2021) · Zbl 1507.74571 |
| [46] | Xia, L.; Breitkopf, P., Design of materials using topology optimization and energy-based homogenization approach in matlab, Struct. Multidiscip. Optim., 52, 6, 1229-1241 (2015) · doi:10.1007/s00158-015-1294-0 |
| [47] | Rozvany, GIN, A critical review of established methods of structural topology optimization, Struct. Multidiscip. Optim., 37, 3, 217-237 (2009) · Zbl 1274.74004 · doi:10.1007/s00158-007-0217-0 |
| [48] | Zhang, Y.; Xiao, M.; Li, H.; Gao, L., Topology optimization of material microstructures using energy-based homogenization method under specified initial material layout, J. Mech. Sci. Technol., 33, 2, 677-693 (2019) · doi:10.1007/s12206-019-0123-6 |
| [49] | Svanberg, K., The method of moving asymptotes-a new method for structural optimization, Int. J. Numer. Meth. Eng., 24, 2, 359-373 (1987) · Zbl 0602.73091 · doi:10.1002/nme.1620240207 |
| [50] | Tyler E, Bruns and Daniel A, Tortorelli: Topology optimization of non-linear elastic structures and compliant mechanisms. Comp. Methods Appl. Mech. Eng. 190(26-27):3443-3459 (2001) · Zbl 1014.74057 |
| [51] | Adam, Paszke, Sam, Gross, Francisco, Massa, Adam, Lerer, James, Bradbury, Gregory, Chanan, Trevor, Killeen, Zeming, Lin, Natalia, Gimelshein, Luca, Antiga, Alban, Desmaison, Andreas, Kopf, Edward, Yang, Zachary, DeVito, Martin, Raison, Alykhan, Tejani, Sasank, Chilamkurthy, Benoit, Steiner, Lu Fang, Junjie, Bai, and Soumith, Chintala. Pytorch: An imperative style, high-performance deep learning library. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d’Alché-Buc, E. Fox, and R. Garnett, editors, Advances in Neural Information Processing Systems 32, pages 8024-8035. Curran Associates, Inc., (2019). URL http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf |
| [52] | Chandrasekhar, A.; Sridhara, S.; Suresh, K., Auto: a framework for automatic differentiation in topology optimization, Struct. Multidiscip. Optim., 64, 6, 4355-4365 (2021) · doi:10.1007/s00158-021-03025-8 |
| [53] | Topology optimization codes written in python. https://www.topopt.mek.dtu.dk/Apps-and-software/Topology-optimization-codes-written-in-Python. Accessed: (2022)-06-25 |
| [54] | Erik, Andreassen, Anders Clausen, Mattias Schevenels, Boyan S Lazarov, and Ole Sigmund: Efficient topology optimization in matlab using 88 lines of code. Structural and Multidisciplinary Optimization, 43 (1):1-16 (2011) · Zbl 1274.74310 |
| [55] | SIMULIA. Abaqus (2020) |
| [56] | MATLAB. version R2021a. The MathWorks Inc., Natick, Massachusetts (2021) |
| [57] | Jeremy Yu, Lu Lu, Xuhui Meng, and George Em. Karniadakis: Gradient-enhanced physics-informed neural networks for forward and inverse pde problems. Comp. Methods Appl. Mech. Eng. 393:114823 (2022) · Zbl 1507.65217 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
