基于可用度模型的故障预测与健康管理方法  

Method of prognostics and health management based on availability model

作　　者：王昊天[1] 石健[2] 

机构地区：[1]北京航空航天大学电子信息工程学院,北京100191 [2]北京航空航天大学自动化科学与电气工程学院,北京100191

出　　处：《系统工程与电子技术》2010年第12期2584-2589,共6页Systems Engineering and Electronics

基　　金：中国国防重点基金(51419020404HK0150);国家自然科学基金(60879024)资助课题

摘　　要：为了将故障预测与健康管理(prognostics and health management,PHM)技术应用到工程实践中,提出了基于可用度模型的PHM方法。首先通过广义随机Petri网(generalized stochastic Petri nets,GSPN)和连续马尔科夫链(continuous time Markov chain,CTMC)建立基本单元的软硬件可用度模型和健康状态转换图,通过求解微分方程得到基本单元软硬件的可用度数值。然后综合软硬件之间的故障相关性建立基本单元的完整可用度模型,并利用事件调度仿真机制得到其可用度的解。最后将基本单元故障模型同通用的可修系统稳态可用度模型对比,得到"可用度-故障率-维修率"形式的PHM计算模型,并以此作为工程应用中PHM分析的有效手段。To apply prognostics and health management（PHM） technology in realistic engineering,a method of PHM based on availability model is proposed.Firstly,the hardware and software model of the basic unit and its healthy state transition diagram are established by generalized stochastic Petri nets（GSPN） and continuous time Markov chain（CTMC）.The availability of the basic unit could be calculated by sloving differential equations.Combining the relationship of the hardware and software faults,the unit availability model is obtained.Based on event scheduling simulation,the result of the basic unit availability is achieved.Finally,a comparison is made between the failure model of the basic uint and the model of repairable system theory,the ＂availability-failure rate-maintenance rate＂ calculation model is obtained and is applied for PHM analysis in engineering.

关 键 词：故障预测与健康管理 广义随机PETRI网 连续马尔可夫链 可用度 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置] V215.7[自动化与计算机技术—控制科学与工程]