%0 Journal Article
%T 用前馈神经网络解稳态对流扩散方程<br>Feedforward Neural Networks for Solving Steady-State Convection Diffusion Equation
%A 孟柳君
%A 成蓉华
%J Advances in Applied Mathematics
%P 28-32
%@ 2324-8009
%D 2022
%I Hans Publishing
%R 10.12677/AAM.2022.111005
%X 本文主要用前馈神经网络求解一维稳态对流扩散方程，因方程中含有一个小参数，所以用传统方法不容易达到理想效果，本文通过构造神经网络模拟方程并与其精确解作对比，选取ε={0.1,0.01}借助软件进行模拟，计算其误差，做出图像。<br />
This paper mainly uses feedforward neural network to solve one-dimensional steady-state convection diffusion equation. The equation contains a small parameter, so the traditional method cannot achieve the ideal effect. This paper constructs the neural network simulation equation and then makes comparison with its accurate solution. This paper chooses parameter <span>ε={0.1,0.01}</span>, uses computer software to simulate, calculates its error, and makes an image.
%K 前馈神经网络，偏微分方程<br>Feedforward Neural Networks
%K Partial Differential Equation
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=47813