Abstract
This paper proposed a new topology optimization method based on geometry deep learning. The density distribution in design domain is described by deep neural networks. Compared to traditional density-based method, using geometry deep learning method to describe the density distribution function can guarantee the smoothness of the boundary and effectively overcome the checkerboard phenomenon. The design variables can be reduced phenomenally based on deep learning representation method. The numerical results for three different kernels including the Gaussian, Tansig, and Tribas are compared. The structural complexity can be directly controlled through the architectures of the neural networks, and minimum length is also controllable for the Gaussian kernel. Several 2-D and 3-D numerical examples are demonstrated in detail to demonstrate the effectiveness of proposed method from minimum compliance to stress-constrained problems.
Similar content being viewed by others
References
Bendsøe MP, Kikuchi N (1988) Generating optimal topologies in structural design using a homogenization method. Comput Methods Appl Mech Eng 71(2):197–224
Wang MY, Wang X, Guo D (2003) A level set method for structural topology optimization. Comput Methods Appl Mech Eng 192(1–2):227–246
Guo X, Zhang W, Zhong W (2014) Doing topology optimization explicitly and geometrically—a new moving morphable components based framework. J Appl Mech 81(8):081009
Sigmund O, Bondsgc M (2003) Topology optimization. State-of-the-art future perspectives. Technical University of Denmark, Copenhagen
Lazarov BS, Wang F, Sigmund O (2016) Length scale and manufacturability in density-based topology optimization. Arch Appl Mech 86(1–2):189–218
Zhang W, Li D, Zhou J, Du Z, Li B, Guo X (2018) A moving morphable void (MMV)-based explicit approach for topology optimization considering stress constraints. Comput Methods Appl Mech Eng 334:381–413
Zhang W et al (2018) Topology optimization with multiple materials via moving morphable component (MMC) method. Int J Numer Methods Eng 113(11):1653–1675
Zhang W et al (2017) Explicit three dimensional topology optimization via moving morphable void (MMV) approach. Comput Methods Appl Mech Eng 322:590–614
Norato J, Bell B, Tortorelli D (2015) A geometry projection method for continuum-based topology optimization with discrete elements. Comput Methods Appl Mech Eng 293:306–327
Zhang S, Gain AL, Norato JA (2017) Stress-based topology optimization with discrete geometric components. Comput Methods Appl Mech Eng 325:1–21
Watts S, Tortorelli DA (2017) A geometric projection method for designing three-dimensional open lattices with inverse homogenization. Int J Numer Methods Eng 112(11):1564–1588
White DA, Stowell ML, Tortorelli DA (2018) Toplogical optimization of structures using Fourier representations. Struct Multidiscipl Optim 58(3):1205–1220
Gao J, Gao L, Luo Z, Li P (2019) Isogeometric topology optimization for continuum structures using density distribution function. Int J Numer Methods Eng 119(10):991–1017
Gulian M, Raissi M, Perdikaris P, Karniadakis G (2019) Machine learning of space-fractional differential equations. SIAM J Sci Comput 41(4):A2485–A2509
Raissi M, Perdikaris P, Karniadakis GE (2019) Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J Comput Phys 378:686–707
Raissi M, Karniadakis GE (2018) Hidden physics models: machine learning of nonlinear partial differential equations. J Comput Phys 357:125–141
Raissi M, Wang Z, Triantafyllou MS, Karniadakis GE (2019) Deep learning of vortex-induced vibrations. J Fluid Mech 861:119–137
Raissi M, Perdikaris P, Karniadakis GE (2017) Machine learning of linear differential equations using Gaussian processes. J Comput Phys 348:683–693
Alber M et al. (2019) Multiscale modeling meets machine learning: what can we learn? arXiv:.11958
Mao Z, Jagtap AD, Karniadakis GE (2020) Physics-informed neural networks for highspeed flows. Comput Methods Appl Mech Eng 360:112789
Zhang D, Lu L, Guo L, Karniadakis GE (2019) Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems. J Comput Phys 397:108850
Meng X, Li Z, Zhang D, Karniadakis GE (2019) PPINN: parareal physics-informed neural network for time-dependent PDEs. arXiv:.10145
Yang L et al. (2019) Highly-scalable, physics-informed GANs for learning solutions of stochastic PDEs. arXiv:.13444
Lei X, Liu C, Du Z, Zhang W, Guo X (2019) Machine learning-driven real-time topology optimization under moving morphable component-based framework. J Appl Mech 86(1):011004
Oh S, Jung Y, Kim S, Lee I, Kang N (2019) Deep generative design: integration of topology optimization and generative models. J Mech Des 141(11):111405
White DA, Arrighi WJ, Kudo J, Watts SE (2019) Multiscale topology optimization using neural network surrogate models. Comput Methods Appl Mech Eng 346:1118–1135
Bengio Y, Courville A, Vincent P (2013) Representation learning: a review and new perspectives. IEEE Trans Pattern Anal 35(8):1798–1828
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436
Qi CR, Su H, Mo K, Guibas LJ (2017) Pointnet: deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 652–660
Litany O, Bronstein A, Bronstein M, Makadia A (2018) Deformable shape completion with graph convolutional autoencoders. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1886–1895
Park JJ, Florence P, Straub J, Newcombe R, Lovegrove S (2019) Deepsdf: learning continuous signed distance functions for shape representation. arXiv:.05103
Lorensen WE, Cline HE (1987) Marching cubes: a high resolution 3D surface construction algorithm. ACM Siggraph Comput Graph 21(4):163–169
Zhou P, Du J, Lü Z (2018) A generalized DCT compression based density method for topology optimization of 2D and 3D continua. Comput Methods Appl Mech Eng 334:1–21
Luo Y, Bao J (2019) A material-field series-expansion method for topology optimization of continuum structures. Comput Struct 225:106122
White DA, Choi Y, Kudo J (2019) A dual mesh method with adaptivity for stress-constrained topology optimization. Struct Multidiscipl Optim 61:749–762
Luo Y, Xing J, Kang Z (2020) Topology optimization using material-field series expansion and Kriging-based algorithm: an effective non-gradient method. Comput Methods Appl Mech Eng 364:112966
Benkő P, Martin RR, Várady T (2001) Algorithms for reverse engineering boundary representation models. Comput Aided Des 33(11):839–851
Hart JC (1998) Morse theory for implicit surface modeling. In: Hart JC (ed) Mathematical visualization. Springer, Berlin, pp 257–268
Gomes A, Voiculescu I, Jorge J, Wyvill B, Galbraith C (2009) Implicit curves and surfaces: mathematics, data structures and algorithms. Springer, Berlin
Ucicr T (1992) Feature-based image metamorphosis. Comput Graph 26:2
Li Q, Hong Q, Qi Q, Ma X, Han X, Tian J (2018) Towards additive manufacturing oriented geometric modeling using implicit functions. Vis Comput Ind Biomed Art 1(1):1–16
Yoo DJ (2011) Porous scaffold design using the distance field and triply periodic minimal surface models. Biomaterials 32(31):7741–7754
Turk G, Levoy M (1994) Zippered polygon meshes from range images. In: Proceedings of the 21st annual conference on computer graphics and interactive techniques, pp 311–318
Tripathi Y, Shukla M, Bhatt AD (2019) Implicit-function-based design and additive manufacturing of triply periodic minimal surfaces scaffolds for bone tissue engineering. J Mater Eng Perform 28(12):7445–7451
Goodfellow I et al. (2014) Generative adversarial nets. In: Advances in neural information processing systems, pp 2672–2680
Dai A, Ruizhongtai Qi C, Nießner M (2017) Shape completion using 3D-encoder-predictor cnns and shape synthesis. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 5868–5877
Tan SM, Michael L (1995) Reducing data dimensionality through optimizing neural network inputs. AIChE J 41(6):1471–1480
Schoen AH (1970) Infinite periodic minimal surfaces without self-intersections. National Aeronautics and Space Administration
Eldan R, Shamir O (2016) The power of depth for feedforward neural networks. In: Conference on learning theory, pp 907–940
Lin HW, Tegmark M, Rolnick D (2017) Why does deep and cheap learning work so well? J Stat Phys 168(6):1223–1247
Liang S, Srikant R (2016) Why deep neural networks for function approximation? arXiv preprint arXiv:1610.04161
Telgarsky M (2015) Representation benefits of deep feedforward networks. arXiv:.08101
Mhaskar HN, Poggio T (2016) Deep vs. shallow networks: an approximation theory perspective. Anal Appl 14(06):829–848
Matsugu M, Mori K, Mitari Y, Kaneda YJNN (2003) Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Netw 16(5–6):555–559
Mhaskar H, Liao Q, Poggio T (2016) Learning functions: when is deep better than shallow. arXiv:.00988
Park JJ, Florence P, Straub J, Newcombe R, Lovegrove S (2019) DeepSDF: learning continuous signed distance functions for shape representation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 165–174
Vogl TP, Mangis J, Rigler A, Zink W, Alkon D (1988) Accelerating the convergence of the back-propagation method. Biol Cybern 59(4–5):257–263
Pujol J (2007) The solution of nonlinear inverse problems and the Levenberg–Marquardt method. Geophysics 72(4):W1–W16
Biegler LT, Conn AR, Coleman TF, Santosa FN (1997) Large-scale optimization with applications: optimal design and control. Springer, Berlin
Andreassen E, Clausen A, Schevenels M, Lazarov BS, Sigmund O (2011) Efficient topology optimization in MATLAB using 88 lines of code. Struct Multidiscipl Optim 43(1):1–16
Lee E, James KA, Martins JR (2012) Stress-constrained topology optimization with design-dependent loading. Struct Multidiscipl Optim 46(5):647–661
Picelli R, Townsend S, Brampton C, Norato J, Kim H (2018) Stress-based shape and topology optimization with the level set method. Comput Methods Appl Mech Eng 329:1–23
Kiyono C, Vatanabe S, Silva E, Reddy J (2016) A new multi-p-norm formulation approach for stress-based topology optimization design. Compos Struct 156:10–19
Lian H, Christiansen AN, Tortorelli DA, Sigmund O, Aage N (2017) Combined shape and topology optimization for minimization of maximal von Mises stress. Struct Multidiscipl Optim 55(5):1541–1557
Zhou M, Sigmund O (2017) On fully stressed design and p-norm measures in structural optimization. Struct Multidiscipl Optim 56(3):731–736
Cai S, Zhang W (2015) Stress constrained topology optimization with free-form design domains. Comput Methods Appl Mech Eng 289:267–290
Xia L, Zhang L, Xia Q, Shi T (2018) Stress-based topology optimization using bi-directional evolutionary structural optimization method. Comput Methods Appl Mech Eng 333:356–370
Wang MY, Li L (2013) Shape equilibrium constraint: a strategy for stress-constrained structural topology optimization. Struct Multidiscipl Optim 47(3):335–352
Le C, Norato J, Bruns T, Ha C, Tortorelli D, Optimization M (2010) Stress-based topology optimization for continua. Struct Multidiscipl Optim 41(4):605–620
Bradley AM (2013) PDE-constrained optimization and the adjoint method. Technical Report. Stanford University. https://cs.stanford.edu/~ambrad/adjoint_tutorial.pdf
Baydin AG, Pearlmutter BA, Radul AA, Siskind JM (2018) Automatic differentiation in machine learning: a survey. J Mach Learn Res 18(153):5595–5637
Bartholomew-Biggs M, Brown S, Christianson B, Dixon L, Mathematics A (2000) Automatic differentiation of algorithms. J Comput 124(1–2):171–190
Andersson JA, Gillis J, Horn G, Rawlings JB, Diehl M (2019) CasADi: a software framework for nonlinear optimization and optimal control. J Math Program Comput 11(1):1–36
Holmberg E, Torstenfelt B, Klarbring A, Optimization M (2013) Stress constrained topology optimization. Struct Multidiscipl Optim 48(1):33–47
Rahmatalla S, Swan C (2004) A Q4/Q4 continuum structural topology optimization implementation. Struct Multidiscipl Optim 27(1–2):130–135
Girosi F, Jones M, Poggio T (1995) Regularization theory and neural networks architectures. Neural Comput 7(2):219–269
Baudat G, Anouar F (2001) Kernel-based methods and function approximation. In: IJCNN’01. International joint conference on neural networks. Proceedings (Cat. No. 01CH37222), vol. 2. IEEE, pp 1244–1249
Elleuch K, Chaari A (2011) Modeling and identification of hammerstein system by using triangular basis functions. Int J Electr Comput Eng 1:1
Wand MP, Jones MC (1994) Kernel smoothing. Chapman and Hall, London
Yakubovich S, Zayed AI (1997) Handbook of function and generalizedfunction transformations. Academic Press, New York
Carstensen JV, Guest JK (2018) Projection-based two-phase minimum and maximum length scale control in topology optimization. Struct Multidiscipl Optim 58(5):1845–1860
Carstensen JV, Guest JK (2014) New projection methods for two-phase minimum and maximum length scale control in topology optimization. In: 15th AIAA/ISSMO multidisciplinary analysis and optimization conference, p 2297
Guest JK, Smith Genut LC (2010) Reducing dimensionality in topology optimization using adaptive design variable fields. Int J Numer Methods Eng 81(8):1019–1045
Guest JK (2009) Topology optimization with multiple phase projection. Comput Methods Appl Mech Eng 199(1–4):123–135
Guest JK (2009) Imposing maximum length scale in topology optimization. Struct Multidiscipl Optim 37(5):463–473
Guest J, Prevost J (2006) A penalty function for enforcing maximum length scale criterion in topology optimization. In: 11th AIAA/ISSMO multidisciplinary analysis and optimization conference, p 6938
Guest JK, Prévost JH, Belytschko T (2004) Achieving minimum length scale in topology optimization using nodal design variables and projection functions. Int J Numer Methods Eng 61(2):238–254
Lazarov BS, Wang F (2017) Maximum length scale in density based topology optimization. Comput Methods Appl Mech Eng 318:826–844
Zhou M, Lazarov BS, Wang F, Sigmund O (2015) Minimum length scale in topology optimization by geometric constraints. Comput Methods Appl Mech Eng 293:266–282
Wang F, Lazarov BS, Sigmund O (2011) On projection methods, convergence and robust formulations in topology optimization. Struct Multidiscipl Optim 43(6):767–784
Sigmund O (2007) Morphology-based black and white filters for topology optimization. Struct Multidiscipl Optim 33(4–5):401–424
Sigmund O (2009) Manufacturing tolerant topology optimization. Acta Mech Sin 25(2):227–239
Lazarov BS, Sigmund O (2011) Filters in topology optimization based on Helmholtz-type differential equations. Int J Numer Methods Eng 86(6):765–781
Svanberg K (2007) MMA and GCMMA-two methods for nonlinear optimization. Optim Syst Theory 1:1–15
Rozvany GI (2009) A critical review of established methods of structural topology optimization. Struct Multidiscipl Optim 37(3):217–237
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105
Freund Y, Mason L (1999) The alternating decision tree learning algorithm. In: icml, vol 99, pp 124–133
Ho TK (1995) Random decision forests. In: Proceedings of 3rd international conference on document analysis and recognition, vol 1. IEEE, pp 278–282
Acknowledgements
The authors would like to acknowledge the support from National Science Foundation (CMMI-1634261).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Deng, H., To, A.C. Topology optimization based on deep representation learning (DRL) for compliance and stress-constrained design. Comput Mech 66, 449–469 (2020). https://doi.org/10.1007/s00466-020-01859-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00466-020-01859-5