A Novel Fourier Feature Physics‐Informed Neural Networks Based on the Boundary Element Method for Solving Scattering of SH Wave Induced by Complex Topography

ABSTRACT The topographic effects of seismic waves often amplify seismic hazards. Thus, solving the scattering phenomena of seismic waves due to complex topography is a significant challenge in seismology and earthquake engineering. To address this problem efficiently and accurately, we propose a novel method, termed FF PINNs‐BEM, which integrates Fourier feature physics‐informed neural networks (FF PINNs) with the boundary element method (BEM) to solve parameterized frequency‐domain wave equations. The accuracy of the proposed method is verified by three representative topographies, that is, a semi‐cylindrical canyon, a semi‐cylindrical mountain and a V‐shaped canyon, for which analytical solutions are available. Furthermore, we simulate scattered wavefields parameterized by the incidence angle and dimensionless frequency. Transfer learning, leveraging a pre‐trained model, is employed to accelerate the neural network convergence. Experimental results demonstrate that the proposed method achieves high accuracy and efficiency while exhibiting robust convergence and generalization capabilities.