江西科学2025，Vol.43Issue(3)：496-501,6.DOI:10.13990/j.issn1001-3679.2025.03.14
基于物理信息神经网络方法的一类圆环域上稳态热传导侧边值问题数值解
Numerical Solution of A Steady-State Heat Conduction Sideways Problem on A Ring Domain Based on Physics-informed Neural Network Method
摘要
Abstract
This study utilizes the physics-informed neural networks(PINNs)method to solve the steady-state heat conduction equation sideways problem on a ring domain.First,the sideways problem is transformed into an optimal control problem.Then the Poisson equation and its boundary conditions are embedded into the loss function of the neural network,and the L-BFGS optimization algorithm is used to minimize the loss function for numerical solution.Numerical experiments demonstrate that the PINNs method effectively solves such problems,showing superior stability,accuracy,and noise resistance.关键词
物理信息神经网络/稳态热传导/侧边值问题/损失函数Key words
physics-informed neural networks/steady-state heat conduction/sideways problem/loss function分类
天文与地球科学引用本文复制引用
刘唐伟,李柯瑶,欧阳旺林,阮晓晴,颜盯盯..基于物理信息神经网络方法的一类圆环域上稳态热传导侧边值问题数值解[J].江西科学,2025,43(3):496-501,6.基金项目
国家自然科学基金项目(42264007). （42264007）
