Physics-Informed Neural Networks with Hard Constraints for Electromagnetic Scattering Analysis
Hang Li, Jingang Liu, Xiaowei Huang, Xin‐Qing Sheng · 2024
In this paper, we present a novel approach to solving two-dimensional electromagnetic scattering problem using physical-informed neural networks (PINN) enhanced with hard constraints. The proposed method is implemented under the partial differential equations (PDEs) framework, which demonstrates superior accuracy over traditional PINN structure compared to conventional numerical methods.