基于随机过程的风力机状态维护建模

Condition-based maintenance modeling-of wind turbine based on stochastic process

为保证系统安全运行、降低系统维护费用,以随机退化的多部件风力机系统为研究对象,构建了状态维护模型。用非定态Gamma过程和非齐次Poisson过程分别对正常退化和随机冲击两个相互影响的退化过程进行建模,考虑维护准备时间和维护时间以及自报警失效和潜在失效两种失效模式,运用更新定理,建立了基于随机退化的状态维护模型。通过实验分析了不同影响因子下期望成本与检测间隔时间之间的关系,结果表明所提出的建模方法对构建带随机退化的多部件系统维护决策问题模型是有效且实用的。

To guarantee system running safely and reduce the maintenance cost, a condition-based maintenance moael was proposed aiming at the stochastic degradations of multi-component wind turbine. Nowstationary gamma process and non-homogeneous poison process were introduced to model the interactive degradation process of normal and stochastic shock respectively. By combining two failure modes of set up time and maintenance time, self-announced failures and latent failures, a condition-based maintenance model based on stochastic degradation was presented with renewal theory. The relationship between the expected maintenance cost and the inspection intervals under different degradation factors and shock factors was analyzed by experiments, and the result showed that the proposed model was valid and practical {or maintenance decision-making of multi-component systems.