节点文献
热传导有限元的神经计算原理及应用
THE NEURAL NETWORK CALCULATING AND ITS APPLICATION TO FINITE ELEMENT METHOD IN HEAT EXCHANGE
【摘要】 根据热传导有限元分析的特点,把有限元计算问题转化为带约束的非线性优化问题,并给出求解决该问题的改进Hopfield(TH)网络,最后对一个简单温度场神经计算进行数值仿真。仿真结果表明,神经网络能完成有限元模型的求解。
【Abstract】 Based on analyzing characteristic of heat exchange with finite element method, the paper transforms the finite element computing problem into a nonlinear optimizing one with some restriction, and solves it using the Hopfield neural network. The result to simulate a simple temperature field with the method shows that the neural network can be applied to finite element model’s resolving.
【关键词】 神经网络;
热传导;
有限元;
仿真;
【Key words】 neural network; heat exchange; finite element method; simulation;
【Key words】 neural network; heat exchange; finite element method; simulation;
- 【文献出处】 物探化探计算技术 ,Computing Techniques For Giophysical and Geochenical Exploration , 编辑部邮箱 ,2002年04期
- 【分类号】TP183
- 【被引频次】6
- 【下载频次】142