Abstract
Layout design is encountered in many fields of engineering and science. Those with complex constraints are particularly challenging to solve due to the non-uniqueness of the solution and the difficulties in incorporating the constraints into the conventional optimization-based methods. In this paper, we propose a design method based on the recently developed machine learning technique, variational autoencoder (VAE). We utilize the learning capability of the VAE to learn the constraints and the generative capability of the VAE to generate design candidates that automatically satisfy all the constraints. As such, no constraints need to be imposed during the design stage. In addition, we show that the VAE network is also capable of learning the underlying physics of the design problem, leading to an efficient design tool that does not need any physical simulation once the network is constructed. We demonstrated the performance of the method on two cases: inverse design of surface diffusion–induced morphology change and mask design for optical microlithography.
Similar content being viewed by others
References
Agrawal BM, Mengüç MP (1991) Forward and inverse analysis of single and multiple scattering of collimated radiation in an axisymmetric system. Int J Heat Mass Transf 34(3):633–647
Bendsøe MP (1989) Optimal shape design as a material distribution problem. Structural Optimization 1(4):193–202
Bendsøe MP, Sigmund O (1999) Material interpolation schemes in topology optimization. Arch Appl Mech 69(9–10):635–654
Bessa M, Bostanabad R, Liu Z, Hu A, Apley DW, Brinson C, Chen W, Liu WK (2017) A framework for data-driven analysis of materials under uncertainty: countering the curse of dimensionality. Comput Methods Appl Mech Eng 320:633–667
Burger M, Osher SJ (2005) A survey in mathematics for industry - a survey on level set methods for inverse problems and optimal design. Eur J Appl Math 16:263–301. https://doi.org/10.1017/s0956792505006182
Byrd RH, Lu P, Nocedal J, Zhu C (1995) A limited memory algorithm for bound constrained optimization. SIAM J Sci Comput 16(5):1190–1208
Cang R, Li H, Yao H, Jiao Y, Ren Y (2018) Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model. Comput Mater Sci 150:212–221
Carleo G, Troyer M (2017) Solving the quantum many-body problem with artificial neural networks. Science 355(6325):602–606
Chen W, Jeyaseelan A, Fuge M (2018) Synthesizing designs with inter-part dependencies using hierarchical generative adversarial networks. In: ASME 2018 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, ASME, Quebec City, Canada
Chen-Wiegart Y-cK, Wang S, Chu YS, Liu W, McNulty I, Voorhees PW, Dunand DC (2012) Structural evolution of nanoporous gold during thermal coarsening. Acta Mater 60(12):4972–4981
Cristoforetti M, Jurman G, Nardelli AI, Furlanello C (2017) Towards meaningful physics from generative models. arXiv preprint arXiv:1705.09524
Doersch C (2016) Tutorial on variational autoencoders. arXiv preprint arXiv:1606.05908
Egorov AA (2004) Inverse problem of theory of the laser irradiation scattering in two-dimensional irregular integrated optical waveguide in the presence of statistic noise. Laser Phys Lett 2(2):77
Forest CE, Allen MR, Stone PH, Sokolov AP (2000) Constraining uncertainties in climate models using climate change detection techniques. Geophys Res Lett 27(4):569–572
Forest CE, Stone PH, Sokolov AP, Allen MR, Webster MD (2002) Quantifying uncertainties in climate system properties with the use of recent climate observations. Science 295(5552):113–117
Franca FHR, Ezekoye OA, Howell JR (2001) Inverse boundary design combining radiation and convection heat transfer. Journal of Heat Transfer-Transactions of the Asme 123(5):884–891. https://doi.org/10.1115/1.1388298
Guo T, Lohan DJ, Cang R, Ren MY, Allison JT (2018) An indirect design representation for topology optimization using variational autoencoder and style transfer. In: 2018 AIAA/ASCE/AHS/ASC structures, structural dynamics, and materials conference. p. 0804
Howell JR, Ezekoye OA, Morales JC (2000) Inverse design model for radiative heat transfer. Journal of Heat Transfer-Transactions of the ASME 122(3):492–502. https://doi.org/10.1115/1.1288774
Jain A, Bollinger JA, Truskett TM (2014) Inverse methods for material design. AICHE J 60(8):2732–2740. https://doi.org/10.1002/aic.14491
Jiang W, Bao W, Thompson CV, Srolovitz DJ (2012) Phase field approach for simulating solid-state dewetting problems. Acta Mater 60(15):5578–5592
Kim J, Song J, Kim K, Kim S, Song J, Kim N, Khan MF, Zhang L, Sader JE, Park K (2016) Hollow microtube resonators via silicon self-assembly toward Subattogram mass sensing applications. Nano Lett 16(3):1537–1545
Kinga D, Adam JB (2015) A method for stochastic optimization. In: International Conference on Learning Representations (ICLR)
Kingma DP, Welling M (2013) Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114
Kingma DP, Mohamed S, Rezende DJ, Welling M (2014) Semi-supervised learning with deep generative models. In: Advances in neural information processing systems. pp. 3581–3589
Kirsch U (2003) A unified reanalysis approach for structural analysis, design, and optimization. Struct Multidiscip Optim 25(2):67–85. https://doi.org/10.1007/s00158-002-0269-0
Krish S (2011) A practical generative design method. Comput Aided Des 43(1):88–100
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems. pp. 1097–1105
Lazarov BS, Wang F, Sigmund O (2016) Length scale and manufacturability in density-based topology optimization. Arch Appl Mech 86(1–2):189–218
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444
Lee AA, Münch A, Süli E (2016) Sharp-Interface limits of the Cahn--Hilliard equation with degenerate mobility. SIAM J Appl Math 76(2):433–456
Li X, Yang Z, Brinson LC, Choudhary A, Agrawal A, Chen W (2018) A deep adversarial learning methodology for designing microstructural material systems. In: ASME 2018 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference 2018, Paper No. DETC2018-85633, pp. V02BT03A008. American Society of Mechanical Engineers
Lohan DJ, Dede EM, Allison JT (2017) Topology optimization for heat conduction using generative design algorithms. Struct Multidiscip Optim 55(3):1063–1077
Mlinar V (2015) Utilization of inverse approach in the design of materials over nano- to macro-scale. Ann Phys-Berlin 527(3–4):187–204. https://doi.org/10.1002/andp.201400190
Mohammadi B, Pironneau O (2004) Shape optimization in fluid mechanics. Annu Rev Fluid Mech 36:255–279. https://doi.org/10.1146/annurev.fluid.36.050802.121926
Mullins WW (1957) Theory of thermal grooving. J Appl Phys 28(3):333–339. https://doi.org/10.1063/1.1722742
Nakshatrala P, Tortorelli D (2015) Topology optimization for effective energy propagation in rate-independent elastoplastic material systems. Comput Methods Appl Mech Eng 295:305–326
Osher SJ, Santosa F (2001) Level set methods for optimization problems involving geometry and constraints: I. Frequencies of a two-density inhomogeneous drum. J Comput Phys 171(1):272–288
Poonawala A, Milanfar P (2007) Mask design for optical microlithography - an inverse imaging problem. IEEE Trans Image Process 16(3):774–788. https://doi.org/10.1109/tip.2006.891332
Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434
Raissi M, Perdikaris P, Karniadakis GE (2017a) Physics informed deep learning (part II): data-driven discovery of nonlinear partial differential equations. arXiv preprint arXiv:1711.10566
Raissi M, Perdikaris P, Karniadakis GE (2017b) Physics informed deep learning (part I): data-driven solutions of nonlinear partial differential equations. arXiv preprint arXiv:1711.10561
Rezende DJ, Mohamed S, Wierstra D (2014) Stochastic backpropagation and approximate inference in deep generative models. arXiv preprint arXiv:1401.4082
Rocchetto A, Grant E, Strelchuk S, Carleo G, Severini S (2017) Learning hard quantum distributions with variational autoencoders. arXiv preprint arXiv:1710.00725
Sayegh SI, Saleh BE (1983) Image design: generation of a prescribed image at the output of a band-limited system. IEEE Trans Pattern Anal Mach Intell (4):441–445
Sigmund O, Maute K (2013) Topology optimization approaches a comparative review. Struct Multidiscip Optim 48(6):1031–1055. https://doi.org/10.1007/s00158-013-0978-6
Sohn K, Lee H, Yan X (2015) Learning structured output representation using deep conditional generative models. In: Advances in neural information processing systems. pp. 3483–3491
Sosnovik I, Oseledets I (2017) Neural networks for topology optimization. arXiv preprint arXiv:1709.09578
Vaz L, Hinton E (1995) FE-shape sensitivity of elastoplastic response. Struct Multidiscip Optim 10(3):231–238
Villanueva CH, Maute K (2017) CutFEM topology optimization of 3D laminar incompressible flow problems. Comput Methods Appl Mech Eng 320:444–473
Wang K, Sun W (2018) A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning. Comput Methods Appl Mech Eng 334:337–380
Wang S, Wang MY (2006) Radial basis functions and level set method for structural topology optimization. Int J Numer Methods Eng 65(12):2060–2090
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
Wetzel SJ (2017) Unsupervised learning of phase transitions: from principle component analysis to variational autoencoders. arXiv preprint arXiv:1703.02435
Yu Y, Hur T, Jung J, Jang IG (2018a) Deep learning for determining a near-optimal topological design without any iteration. Struct Multidiscip Optim:1–13
Yu Y, Hur T, Jung J (2018b) Deep learning for topology optimization design. arXiv preprint arXiv:1801.05463
Zeng F, Wong M (2015) A self-scanned active-matrix tactile sensor realized using silicon-migration technology. J Microelectromech Syst 24(3):677–684
Zeng F, Luo Y, Yobas L, Wong M (2013) Self-formed cylindrical microcapillaries through surface migration of silicon and their application to single-cell analysis. J Micromech Microeng 23(5):055001
Zhang W, Zhou Y, Zhu J (2017) A comprehensive study of feature definitions with solids and voids for topology optimization. Comput Methods Appl Mech Eng 325:289–313
Zhou Y, Zhang W, Zhu J, Xu Z (2016) Feature-driven topology optimization method with signed distance function. Comput Methods Appl Mech Eng 310:1–32
Zhou M, Lazarov BS, Sigmund O (2017) Topology optimization for optical microlithography with partially coherent illumination. Int J Numer Methods Eng 109(5):631–647
Funding
This work is supported by the Hong Kong Research Grants Council under Competitive Earmarked Research Grant No. 16212318.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Responsible Editor: Nestor V Queipo
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published mapsand institutional affiliations.
Surface diffusion solver
Surface diffusion solver
A phase-field model is used to describe interface evolution caused by surface diffusion, which is shown as follows:
In these equations, ϕ denotes the order parameter for the phases of the system: ϕ = − 1 represents a void phase and ϕ = 1 represents a solid phase; μ is the chemical potential; ε is a parameter controlling the thickness of the interface; f(ϕ) is the bulk free energy; and M(ϕ) is the mobility. As ε → 0, this model converges to Mullins sharp interface model (7) for surface diffusion, see (Lee et al. 2016).
The phase-field equations are solved numerically on a periodic domain by using an operator-splitting-based, quasi-spectral, semi-implicit time-stepping scheme. The semi-implicit scheme for (8) can be written in the following form (Jiang et al. 2012),
where B and S are two positive stabilizing parameters chosen to guarantee the numerical stability. Performing Fourier transform on (12), we obtain,
where \( \widehat{\phi}(k) \) is the Fourier transform of ϕ(x), and F{∙} stands for Fourier transform.
Then, we can derive the following explicit time-stepping scheme in the spectrum space,
Rights and permissions
About this article
Cite this article
Zhang, Y., Ye, W. Deep learning–based inverse method for layout design. Struct Multidisc Optim 60, 527–536 (2019). https://doi.org/10.1007/s00158-019-02222-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00158-019-02222-w