论文部分内容阅读
提出结构零件可靠度计算的人工神经网络—蒙特卡罗仿真算法。针对结构零件可靠度计算问题,利用有限元计算方法,得到一组结构输入基本变量及输出响应数据,采用人工神经网络后向传播BP模型,建立结构的输入—输出关系,在该模型上,用Monte Carlo 方法进行结构零件可靠度仿真计算。算例表明,该方法正确可行
The artificial neural network (Monte Carlo simulation algorithm) is proposed to calculate the reliability of structural parts. Aiming at the problem of reliability calculation of structural components, a set of structural input basic variables and output response data are obtained by using finite element method. BP neural network is used to propagate BP model backwards and establish the structure of input-output relationship. In this model, Monte Carlo method for structural component reliability simulation. The example shows that this method is correct and feasible