Litcius/Paper detail

Physics-Informed Supervised Residual Learning for Electromagnetic Modeling

Tao Shan, Jinhong Zeng, Xiaoqian Song, Rui Guo, Maokun Li, Fan Yang, Shenheng Xu

2023IEEE Transactions on Antennas and Propagation43 citationsDOI

Abstract

In this study, physics-informed supervised residual learning (PhiSRL) is proposed to enable an effective, robust, and general deep learning framework for 2-D electromagnetic (EM) modeling. Based on the mathematical connection between the fixed-point iteration method and the residual neural network (ResNet), PhiSRL aims to solve a system of linear matrix equations. It applies convolutional neural networks (CNNs) to learn updates of the solution with respect to the residuals. Inspired by the stationary and nonstationary iterative scheme of the fixed-point iteration method, stationary and nonstationary iterative physics-informed ResNets (SiPhiResNet and NiPhiResNet) are designed to solve the volume integral equation (VIE) of EM scattering. The effectiveness and universality of PhiSRL are validated by solving VIE of lossless and lossy scatterers with the mean squared errors (MSEs) converging to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 10^{-4}$ </tex-math></inline-formula> (SiPhiResNet) and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 10^{-7}$ </tex-math></inline-formula> (NiPhiResNet). Numerical results further verify the generalization ability of PhiSRL.

Topics & Concepts

ResidualArtificial neural networkLossy compressionNotationApplied mathematicsGeneralizationConvolutional neural networkComputer scienceArtificial intelligenceAlgorithmMathematicsMathematical analysisArithmeticElectromagnetic Simulation and Numerical MethodsElectromagnetic Scattering and AnalysisNumerical methods in engineering