摘要
基于Akaike信息准则(AIC)、Bayesian信息准则(BIC)及故障数据拟合的均方根误差(RMSE),提出了可修系统可靠性分析的非齐次泊松过程模型的选择方法,利用最大似然估计的渐近对数正态分布特性,用Fisher信息矩阵法给出了边界强度过程模型参数及系统可靠性指标的点估计及区间估计,分析了多台数控机床时间截尾的故障过程.结果表明,对于维修频繁的性能恶化数控机床,边界强度故障模型适合于其可靠性评估.
Based on Akaike information criterion,Bayesian information criterion and the root-mean-square error of the fitting failure data,the best non-homogeneous Poisson process model of reliability analysis for repairable system was proposed.The point maximum likelihood and interval estimators of the parameters,as well as the reliability indices of the bounded intensity process model were given using the asymptotic lognormal distribution of the maximum likelihood estimation and the Fisher information matrix method.The failure data with time truncation of multiple NC machine tools were analyzed.The results show that the bounded intensity process is suitable for reliability assessment of deterioration machine tools with frequent maintenance actions.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2012年第10期1622-1626,1631,共6页
Journal of Shanghai Jiaotong University
基金
国家科技重大专项资助项目(2009ZX04014-022)
关键词
边界强度过程
可修系统
最大似然估计
数控机床
可靠性评估
bounded intensity process
repairable system
maximum likelihood estimation
NC machine tool
reliability assessment
