论文部分内容阅读
针对大多可靠性工程问题中机构极限状态函数为隐式的情况,提出了一种基于极限学习机(ELM)回归近似极限状态方程的可靠性及灵敏度分析的新方法.通过极限学习机与蒙特卡洛法相结合,利用极限学习机快速学习的能力,将复杂或隐式极限状态方程近似等价为显式极限状态方程,运用蒙特卡洛法计算出机构的失效概率,然后由高精度的显式极限状态方程进行各随机变量参数的灵敏度分析.该方法采用了基于单隐层前馈神经网络极限学习算法,因而在拟合非线性极限状态方程上表现优越,计算精度和效率高.最后以某型起落架中可折支撑锁机构为对象,进行了机构的可靠性及敏感度分析.结果表明:该方法具有高精度和高效率的优点,在工程应用上具有一定的价值.
Aiming at the situation that the limit state function of institutions is implicit in most reliability engineering problems, a new method of reliability and sensitivity analysis based on the limit state-of-learning equation (ELM) regression is proposed. By means of extreme learning machine and Monte Carlo Lofa combination, the use of extreme learning machine learning ability quickly, the complex or implicit limit state equation approximately equal to the explicit limit state equation, the use of Monte Carlo method to calculate the failure probability of institutions, and then by the high-precision explicit The limit state equation is used to conduct the sensitivity analysis of the parameters of each random variable.The method adopts the limit learning algorithm based on single hidden layer feedforward neural network and thus excels in fitting the nonlinear limit state equation with high accuracy and efficiency.At last, Type undercarriage, the reliability and sensitivity of the mechanism are analyzed.The results show that the method has the advantages of high precision and high efficiency, and has certain value in engineering application.