PHYSICS-INFORMED NEURAL NETWORKS WITH HARD CONSTRAINTS FOR INVERSE DESIGN

L. Lu, R. Pestourie, W. Yao, Z. Wang, F. Verdugo, S.G. Johnson

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

© 2021 Society for Industrial and Applied Mathematics.Inverse design arises in a variety of areas in engineering such as acoustic, mechanics, thermal/electronic transport, electromagnetism, and optics. Topology optimization is an important form of inverse design, where one optimizes a designed geometry to achieve targeted properties parameterized by the materials at every point in a design region. This optimization is challenging, because it has a very high dimensionality and is usually constrained by partial differential equations (PDEs) and additional inequalities. Here, we propose a new deep learning method-physics-informed neural networks with hard constraints (hPINNs)-for solving topology optimization. hPINN leverages the recent development of PINNs for solving PDEs, and thus does not require a large dataset (generated by numerical PDE solvers) for training. However, all the constraints in PINNs are soft constraints, and hence we impose hard constraints by using the penalty method and the augmented Lagrangian method. We demonstrate the effectiveness of hPINN for a holography problem in optics and a fluid problem of Stokes flow. We achieve the same objective as conventional PDE-constrained optimization methods based on adjoint methods and numerical PDE solvers, but find that the design obtained from hPINN is often smoother for problems whose solution is not unique. Moreover, the implementation of inverse design with hPINN can be easier than that of conventional methods because it exploits the extensive deep-learning software infrastructure.
Original languageEnglish
Pages (from-to)B1105-B1132
JournalSIAM Journal on Scientific Computing
Volume43
Issue number6
DOIs
Publication statusPublished - 2021
Externally publishedYes

Funding

\ast Submitted to the journal's Computational Methods in Science and Engineering section February 9, 2021; accepted for publication (in revised form) August 11, 2021; published electronically November 11, 2021. https://doi.org/10.1137/21M1397908 \bfF \bfu \bfn \bfd \bfi \bfn \bfg : This work was supported by the MIT-IBM Watson AI Laboratory (challenge 2415). The work of the fifth author was supported by the Spanish Ministry of Economy and Competitiveness through the ``Severo Ochoa Programme for Centers of Excellence in R\&D (CEX2018-000797-S)."" \dagger Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA 19104 USA ([email protected]). \ddagger Department of Mathematics, Massachusetts Institute of Technology, Cambridge, MA 02139 USA ([email protected], [email protected], [email protected]). \S Laboratory of Ocean Energy Utilization of Ministry of Education, Dalian University of Technology, Dalian, 116024, China (zhicheng [email protected]). \P Centre Internacional de M\ètodes Num\èrics en Enginyeria, Esteve Terradas 5, 08860 Castelldefels, Barcelona, Spain ([email protected]).

FundersFunder number
Ministerio de Economía y CompetitividadCEX2018-000797-S

    Fingerprint

    Dive into the research topics of 'PHYSICS-INFORMED NEURAL NETWORKS WITH HARD CONSTRAINTS FOR INVERSE DESIGN'. Together they form a unique fingerprint.

    Cite this