Multistep asymptotic pre-training strategy based on PINNs for solving steep boundary singular perturbation problems

DOI10.1016/j.cma.2024.117222MaRDI QIDQ6609750

Fujun Cao, Fei Gao, Dongfang Yuan, Junmin Liu

Publication date: 24 September 2024

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

zbMATH Keywords

singularly perturbed problemphysics-informed neural networksasymptotic parameter strategymultiple time pre-training strategy

Mathematics Subject Classification ID

Probabilistic models, generic numerical methods in probability and statistics (65C20) Singular perturbations in context of PDEs (35B25) Initial-boundary value problems for second-order parabolic equations (35K20) Spectral, collocation and related methods for initial value and initial-boundary value problems involving PDEs (65M70)

This page was built for publication: Multistep asymptotic pre-training strategy based on PINNs for solving steep boundary singular perturbation problems