GRIDS-Net: Inverse shape design and identification of scatterers via geometric regularization and physics-embedded deep learning

This study presents a deep learning based methodology for both remote sensing and design of acoustic scatterers. The ability to determine the shape of a scatterer, either in the context of material design or sensing, plays a critical role in many practical engineering problems. This class of inverse problems is extremely challenging due to their high-dimensional, nonlinear, and ill-posed nature. To overcome these technical hurdles, we introduce a geometric regularization approach for deep neural networks (DNN) based on non-uniform rational B-splines (NURBS) and capable of predicting complex 2D scatterer geometries in a parsimonious dimensional representation. Then, this geometric regularization is combined with physics-embedded learning and integrated within a robust convolutional autoencoder (CAE) architecture to accurately predict the shape of 2D scatterers in the context of identification and inverse design problems. An extensive numerical study is presented in order to showcase the remarkable ability of this approach to handle complex scatterer geometries while generating physically-consistent acoustic fields. The study also assesses and contrasts the role played by the (weakly) embedded physics in the convergence of the DNN predictions to a physically consistent inverse design.

[1]  Zhenpei Wang,et al.  Deep-learning-based isogeometric inverse design for tetra-chiral auxetics , 2021, Composite Structures.

[2]  Mou Wang,et al.  Inverse design and experimental verification of an acoustic sink based on machine learning , 2021 .

[3]  M. Sacks,et al.  Simulating the time evolving geometry, mechanical properties, and fibrous structure of bioprosthetic heart valve leaflets under cyclic loading. , 2021, Journal of the mechanical behavior of biomedical materials.

[4]  Markus J. Buehler,et al.  Artificial intelligence and machine learning in design of mechanical materials. , 2021, Materials horizons.

[5]  Steven G. Johnson,et al.  Physics-informed neural networks with hard constraints for inverse design , 2021, SIAM J. Sci. Comput..

[6]  A. Raman,et al.  Global Inverse Design across Multiple Photonic Structure Classes Using Generative Deep Learning , 2020, Advanced Optical Materials.

[7]  A. Wills,et al.  Physics-informed machine learning , 2021, Nature Reviews Physics.

[8]  Xinshuai Zhang,et al.  Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization , 2021, Computer Methods in Applied Mechanics and Engineering.

[9]  Steven G. Johnson,et al.  Active learning of deep surrogates for PDEs: application to metasurface design , 2020, npj Computational Materials.

[10]  Jiaqi Jiang,et al.  Deep neural networks for the evaluation and design of photonic devices , 2020, Nature Reviews Materials.

[11]  A. Massa,et al.  A Multiresolution Contraction Integral Equation Method for Solving Highly Nonlinear Inverse Scattering Problems , 2020, IEEE Transactions on Microwave Theory and Techniques.

[12]  Trevon Badloe,et al.  Deep learning enabled inverse design in nanophotonics , 2020, Nanophotonics.

[13]  Prasanna Balaprakash,et al.  Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders , 2020, Physics of Fluids.

[14]  George Em Karniadakis,et al.  Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations , 2020, Science.

[15]  L. Dal Negro,et al.  Physics-informed neural networks for inverse problems in nano-optics and metamaterials. , 2019, Optics express.

[16]  Steven G. Johnson,et al.  Inverse Designed Metalenses with Extended Depth of Focus , 2019, 2020 Conference on Lasers and Electro-Optics (CLEO).

[17]  Hadi Meidani,et al.  Physics-Driven Regularization of Deep Neural Networks for Enhanced Engineering Design and Analysis , 2018, J. Comput. Inf. Sci. Eng..

[18]  Soumik Sarkar,et al.  A Deep Learning Framework for Design and Analysis of Surgical Bioprosthetic Heart Valves , 2019, Scientific Reports.

[19]  Alán Aspuru-Guzik,et al.  Inverse Design of Solid-State Materials via a Continuous Representation , 2019, Matter.

[20]  P. Perdikaris,et al.  Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks , 2019 .

[21]  Gang Sun,et al.  Application of deep learning based multi-fidelity surrogate model to robust aerodynamic design optimization , 2019, Aerospace Science and Technology.

[22]  S. Mostafa Mousavi,et al.  Unsupervised Clustering of Seismic Signals Using Deep Convolutional Autoencoders , 2019, IEEE Geoscience and Remote Sensing Letters.

[23]  Daniel A. White,et al.  Multiscale topology optimization using neural network surrogate models , 2019, Computer Methods in Applied Mechanics and Engineering.

[24]  Junsuk Rho,et al.  Designing nanophotonic structures using conditional deep convolutional generative adversarial networks , 2019, Nanophotonics.

[25]  Paris Perdikaris,et al.  Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations , 2019, J. Comput. Phys..

[26]  Mengqi Zhang,et al.  Inverse Design of Airfoil Using a Deep Convolutional Neural Network , 2019, AIAA Journal.

[27]  Seokho Kang,et al.  Deep-learning-based inverse design model for intelligent discovery of organic molecules , 2018, npj Computational Materials.

[28]  Steven G. Johnson,et al.  Inverse design of large-area metasurfaces. , 2018, Optics express.

[29]  R. Haftka,et al.  Similarity measures for identifying material parameters from hysteresis loops using inverse analysis , 2018, International Journal of Material Forming.

[30]  Bu-Sung Lee,et al.  Autoencoder-based network anomaly detection , 2018, 2018 Wireless Telecommunications Symposium (WTS).

[31]  Jelena Vucković,et al.  Inverse design in nanophotonics , 2018, Nature Photonics.

[32]  W. Cai,et al.  A Generative Model for Inverse Design of Metamaterials , 2018, Nano letters.

[33]  Sanjeev Kumar,et al.  NURBS-based geometric inverse reconstruction of free-form shapes , 2017, J. King Saud Univ. Comput. Inf. Sci..

[34]  Michael Unser,et al.  Convolutional Neural Networks for Inverse Problems in Imaging: A Review , 2017, IEEE Signal Processing Magazine.

[35]  A. Polman,et al.  Nanophotonics: Shrinking light-based technology , 2015, Science.

[36]  Roland Potthast,et al.  A Survey on Inverse Problems for Applied Sciences , 2013 .

[37]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[38]  H. Fleckenstein,et al.  X-ray diffraction imaging with the Multiple Inverse Fan Beam topology: principles, performance and potential for security screening. , 2012, Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine.

[39]  Yanfei Wang,et al.  Optimization and regularization for computational inverse problems and applications , 2011 .

[40]  Jakob S. Jensen,et al.  Acoustic design by topology optimization , 2008 .

[41]  Mario Bertero,et al.  Inverse problems in biomedical imaging: modeling and methods of solution , 2006 .

[42]  Fred W. Glover,et al.  Simulation optimization: a review, new developments, and applications , 2005, Proceedings of the Winter Simulation Conference, 2005..

[43]  T. Hughes,et al.  Isogeometric analysis : CAD, finite elements, NURBS, exact geometry and mesh refinement , 2005 .

[44]  P. Pinsky,et al.  An application of shape optimization in the solution of inverse acoustic scattering problems , 2004 .

[45]  A. Abubakar,et al.  A General Framework for Constraint Minimization for the Inversion of Electromagnetic Measurements , 2004 .

[46]  Gerd Dobmann,et al.  Robust solutions of inverse problems in electromagnetic non-destructive evaluation , 2002 .

[47]  Aria Abubakar,et al.  Total variation as a multiplicative constraint for solving inverse problems , 2001, IEEE Trans. Image Process..

[48]  J. Samareh.  Survey of Shape Parameterization Techniques for High-Fidelity Multidisciplinary Shape Optimization , 2001 .

[49]  J. Kruth,et al.  NURBS curve and surface fitting for reverse engineering , 1998 .

[50]  Dante Fratta,et al.  Introduction to Discrete Signals and Inverse Problems in Civil Engineering , 1998 .

[51]  R. Macleod,et al.  Short titles: Inverse Problems in Electrocardiology , 2022 .

[52]  Thomas Bäck,et al.  Evolutionary computation: comments on the history and current state , 1997, IEEE Trans. Evol. Comput..

[53]  Les A. Piegl,et al.  The NURBS Book , 1995, Monographs in Visual Communication.

[54]  Ole Sigmund,et al.  Some Inverse Problems in Topology Design of Materials and Mechanisms , 1996 .

[55]  Dara Entekhabi,et al.  Solving the inverse problem for soil moisture and temperature profiles by sequential assimilation of multifrequency remotely sensed observations , 1994, IEEE Trans. Geosci. Remote. Sens..

[56]  H. Mannila,et al.  Computing Discrete Fréchet Distance ∗ , 1994 .

[57]  R. Kress,et al.  Inverse Acoustic and Electromagnetic Scattering Theory , 1992 .

[58]  F. Fahy,et al.  Sound and Structural Vibration: Radiation, Transmission and Response , 1987 .

[59]  F. Semperlotti,et al.  A physics-constrained deep learning based approach for acoustic inverse scattering problems , 2022 .