基于Petri网的复杂设备健康状态退化分析

摘要

针对大型设备结构复杂，健康状态难以评估的问题，以设备最小可监测或可评估单元健康状态为输入，建立了健康退化Petri网模型，利用健康等级信度的变化描述复杂设备全系统各层次单元的健康状态退化情况。首先给出了健康退化Petri网的定义，分析了库所间“冗余”、“共同”、“并发”以及“继发”四种关系形式，用以评估各层次单元健康状态；然后分析了健康状态信度转换为健康度的规则并给出一种线性映射方法；根据健康状态退化趋势特点，在离散灰色模型中引入非线性时变项，提出了基于粒子群优化的拓展非线性时变参数离散灰色模型的健康度预测算法。最后将模型应用于某发射装置的健康状态退化分析，说明了模型的适用性。%Due to the complex structures of large equipment,it is difficult to assess heath conditions of large equipment accurately.To solve this problem,health degradation Petri nets (HDPN)is presented.In this mo-del,degradation information of the minimum monitoring units is taken as the input.The degradation of health conditions is described by the changes of the credibility of health grades.Four types of relationships between places are analyzed to assess health conditions accurately.Then,the transforming rules between health degree and health credibility are proposed.According to the characteristics of the health condition degradation,non-li-near time-varying parameters are introduced in the discrete grey model,which is used to forecast health condi-tions.Finally,health degradation of some launcher is analyzed by using the proposed model.Results show that this model is practicable.