{"id":"arxiv/2606.02335","url":"https://arxiv.org/abs/2606.02335","tags":["cond-mat.mtrl-sci","math-ph","physics.app-ph","physics.comp-ph"],"score":0,"title":"Neural Spectral Element Methods for stiff multiphysics PDEs with electrochemical transport benchmarks","domain":"cond-mat.mtrl-sci","source":"arxiv","authors":["Conrard Giresse Tetsassi Feugmo","David Pankaczy"],"reviews":0,"subject":"cond-mat.mtrl-sci","summary":"The Neural Spectral Element Method (NSEM) evaluates each network only at fixed Legendre-Gauss-Lobatto quadrature nodes and replaces all derivative calls with precomputed spectral differentiation matrices. The resulting deterministic loss enables limited-memory BFGS (L-BFGS) to reach residuals of 10^-9 to 10^-10. A Kosloff-Tal-Ezer coordinate map resolves electrochemical boundary layers, while a mesh-free neural mortar framework couples multi-element domains. On the four-example Poisson-Nernst-Planck (PNP) benchmark of Huang and co-workers, NSEM attains 10^-4 to 10^-7 relative pointwise error…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-06-01","reactions":{}}