Answers are grounded in this paper and cite their sources.
This work presents a deep learning-based framework for the solution of partial differential equations on complex computational domains described with computer-aided design tools. To account for the underlying distribution of the training data caused by spline-based projections from the reference to the physical domain, a variational neural solver equipped with an importance sampling scheme is developed, such that the loss function based on the discretized energy functional obtained after the weak formulation is modified according to the sample distribution. To tackle multipatch domains possibly leading to solution discontinuities, the variational neural solver is additionally combined with a domain decomposition approach based on the discontinuous Galerkin formulation. The proposed neural solver is verified on a toy problem and then applied to a real-world engineering test case, namely, that of electric machine simulation. The numerical results show clearly that the neural solver produces physics-conforming solutions of significantly improved accuracy.
No takes yet. Share an insight, caveat, or question.
Tresckow et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: