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.
出处
《河南科技大学学报（自然科学版）》
CAS
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)
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
U-shaped flexural electro-thermal actuator
Stochastic finite element method
BP neural network
Monte Carlo method
作者简介
尹婷婷（1983-），女，安徽淮南人，硕士生；
高行山（1964-），男，陕西安塞人，教授，博士，主要从事工程力学、微／纳机电系统力学、工程结构优化设计方面的研究工作．
