基于神经网络的电热式微致动器随机有限元分析  

Stochastic Finite Element Analysis of Flexural Electro-thermal Actuator Based on Neural Network

作　　者：尹婷婷[1] 高行山[1,2] 张娟[1,2] 

机构地区：[1]西北工业大学工程力学系,陕西西安710072 [2]大连大学辽宁省智能信息处理重点实验室,辽宁大连116622

出　　处：《河南科技大学学报（自然科学版）》2008年第6期38-41,共4页Journal of Henan University of Science And Technology:Natural Science

基　　金：辽宁省智能信息处理重点实验室开放基金(2005-5)

摘　　要：提出了基于神经网络的随机有限元分析方法,首先利用有限元法对U型电热式微致动器进行理论分析,得出具体结构尺寸对微致动器最大位移的影响,然后通过建立的神经网络来拟合响应与输入之间的关系,根据蒙特卡罗模拟原理获得足够多的样本值对训练后的网络进行误差分析,结果证明,本文提出的分析方法是可行有效的。The random finite element method was proposed based on the BP artificial neural network.First the maximal displacement of the U-shaped flexural electro-thermal actuators whose concrete size affected geometrical structure was solved by the finite element method through the theoretic analysis.The relation between the response and the input was simulated through the stochastic finite element method.The Monte Carlo method was applied to get efficient data of the limit state for reliability analysis and to analyze the error.The result shows that the method of this paper is feasible and effective.

关 键 词：U型电热式微致动器 随机有限元 BP神经网络 蒙特卡罗法 

分 类 号：O316[理学—一般力学与力学基础] TP389.1[理学—力学]