基于拟似然函数的集成仪器可靠性估计  

Integrated instrument reliability estimation based on quasi likelihood function

作　　者：邓昌建[1,2] 

机构地区：[1]成都信息工程学院控制工程系,成都610225 [2]电子科技大学自动化学院,成都611731

出　　处：《仪器仪表学报》2011年第6期1414-1419,共6页Chinese Journal of Scientific Instrument

摘　　要：由于失效分布函数的不可知或难以描述,经典统计估计在处理集成仪器的人为故障失效数据、复杂失效数据和无失效数据时,很难估计出其可靠性参数。结合测量系统的常见失效分布,研究了实验仪器的拟似然函数的引入和常见几种类型的拟似然函数应用方法;作为拟似然函数应用的基础,给出了拟似然函数的近似计算方法与证明。最后讨论了最大连续似然估计与最大分位似然估计(QMPE)的区别,及使用QMPE的方法的步骤。通过实际的测量数据和其可靠性估计,验证了拟似然估计在集成仪器可靠性分析中的可行性。Because of the difficulties in describing and finding failure distribution function,it is difficult to estimate reliability parameters using classical statistics method when dealing with the failure data caused by human,complicated failure data and zero-failure data of integrated instrument.Considering the common failure distribution of measurement system,the introduction of quasi likelihood functions and the application method of commonly used quasi likelihood functions for experimental instrument are discussed.As the basis of the application of quasi likelihood function,the approximate computing method and its proof are presented in the paper.In the next part of the paper,the difference between the CML and QMPE is discussed,and the method and procedure of using QMPE are introduced also.Experiment data and reliability analysis data verify the feasibility of using quasi likelihood function in the reliability analysis of integrated instrument.

关 键 词：可靠性 拟似然函数 失效分布 近似计算 最大分位似然估计 

分 类 号：TP202.1[自动化与计算机技术—检测技术与自动化装置]