威布尔比例风险模型装备状态维修检测间隔期研究

摘要

According to the fact the real life of equipment is from complex factors,based on Weibull Proportional Hazards Model (WPHM), reliability model is established. Considering disadvantage of traditional methods, genetic algorithm is utilized to estimate the values of model parameters. At same time, A few conditional parameters can substitute for original multidimensional variables by principal component analysis with little data information lost. Based on failure risk, test intervals model for Condition Based Maintenance (CBM) is established. The result shows the model works well and effectively.%针对装备的实际寿命受到多种复杂因素影响的情况，以威布尔比例风险模型为基础，建立了反应装备状态的可靠度模型，针对传统参数估计方法的不足，提出了利用遗传算法进行模型参数估计的方法。同时，利用主成分分析法，保证原始数据信息损失最小的情况下，以少数的综合变量取代原有的多维变量，简化了状态参数。以故障风险为决策目标，建立了状态维修检测间隔期决策模型。通过实例，验证了该方法的可行性和有效性。