人工神经网络
有限元法
搭配(遥感)
反问题
解算器
反向
偏微分方程
边界(拓扑)
计算机科学
应用数学
集合(抽象数据类型)
数学优化
边值问题
数学
算法
数学分析
人工智能
几何学
机器学习
物理
程序设计语言
热力学
作者
Santiago Badia,Wei Li,Alberto F. Martı́n
出处
期刊:Cornell University - arXiv
日期:2023-06-09
标识
DOI:10.1016/j.cma.2023.116505
摘要
We propose a general framework for solving forward and inverse problems constrained by partial differential equations, where we interpolate neural networks onto finite element spaces to represent the (partial) unknowns. The framework overcomes the challenges related to the imposition of boundary conditions, the choice of collocation points in physics-informed neural networks, and the integration of variational physics-informed neural networks. A numerical experiment set confirms the framework's capability of handling various forward and inverse problems. In particular, the trained neural network generalises well for smooth problems, beating finite element solutions by some orders of magnitude. We finally propose an effective one-loop solver with an initial data fitting step (to obtain a cheap initialisation) to solve inverse problems.
科研通智能强力驱动
Strongly Powered by AbleSci AI