基于神经网络的机械零部件可靠性优化设计

采用概率约束等价转换的数值逼近法，研究了具有任意分布参数的可靠性优化设计，可以迅速准确地获得优化设计结果。针对具有多失效模式的机械零部件可靠性优化设计，提出了随机模拟-神经网络(MCSNN)方法，模拟得到随机设计变量与机械零部件系统可靠度之间的显性函数表达式，简化了计算过程，同时可以获得较高的计算精度，具有较好的工程实用价值。

Reliability-based Optimization for Mechanical Components Using Neural Network

Two methods were investigated in reliability-based optimization for mechanical components. Firstly, a numerical approach method was used to solve the reliability-based optimization with arbitrary distribution random parameters. By the method, the probabilistic constraints could be transformed into deterministic constraints, and the reliability-based optimal design parameters could be obtained accurately and quickly. Secondly, MCS-NN method was applied in reliability-based optimization for mechanical components with many failure modes. Therefore, the explicit expression between the design parameters and the system reliability was given correctly, the process of reliability-based optimization can be implemented expediently.