一种搜索威布尔MLE形状参数的PSO联合方法

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (4): 988-993.

一种搜索威布尔MLE形状参数的PSO联合方法

罗航, 龙兵, 黄建国, 王厚军

电子科技大学自动化工程学院, 四川, 成都, 610054

收稿日期:2007-11-13 修回日期:2008-06-06 出版日期:2009-04-20 发布日期:2010-01-03
作者简介:罗航(1972- ),男,博士研究生,主要研究方向为故障诊断和可靠性预测.E-mail:luohang2002@163.com
基金资助:
国防基础科研项目(A1420061264);国家自然科学基金(60673011)资助课题

Method of searching Weibull shape parameter under MLE based on united PSO

LUO Hang, LONG Bing, HUANG Jian-guo, WANG Hou-jun

Coll. of Automation Engineering, Univ. of Electronic Science and Technology of China, Chengdu 610054, China

Received:2007-11-13 Revised:2008-06-06 Online:2009-04-20 Published:2010-01-03

摘要/Abstract

摘要： 以两参数威布尔分布的极大似然估计(MLE)方法为背景,首次将粒子群优化(PSO)算法用于形状参数的MLE估计。在详细分析PSO方法特点和实现的基础上,将基于最小二乘估计(LSE)的线性回归方法和PSO相结合,实现了一种联合优化的方法。与此同时,评析了将LSE和PSO相结合后用于两参数威布尔分布的MLE寻优方法。仿真和实际运用表明,基于PSO的联合方法对威布尔分布的MLE形状参数的搜索具有良好的性质,其具体体现为搜索准确、稳定和快捷。

Abstract: A method named particle swarm optimization(PSO) is firstly used for estimating the Weibull shape parameter,which is based on maximum likelihood estimation(MLE) about two-parameter Weibull distribution.The method of linear regression based on least-square estimation(LSE) is combined with PSO,then a kind of united optimization method is presented and corresponding comments are given.Simulation and actual application show that the presented united optimization method used for searching the shape parameter of Weibull distribution based on the method of MLE has good characteristics,namely,the searching is correct,robust and quick.

中图分类号:

TP272⁺.1

罗航, 龙兵, 黄建国, 王厚军. 一种搜索威布尔MLE形状参数的PSO联合方法[J]. Journal of Systems Engineering and Electronics, 2009, 31(4): 988-993.

LUO Hang, LONG Bing, HUANG Jian-guo, WANG Hou-jun. Method of searching Weibull shape parameter under MLE based on united PSO[J]. Journal of Systems Engineering and Electronics, 2009, 31(4): 988-993.

参考文献

[1] Murthy DNP,Bulmer M,Eccleston JA.Weibull model selection for reliability modelling[J].Reliability Engineering & System Safety,2004,86(3):257-267.
[2] Xie M,Goh TN,Tang Y.On changing points of mean residual life and failure rate function for some generalized Weibull distributions[J].Reliability Engineering & System Safety,2004,84 (3):293-299.
[3] Nadarajah S.On the moments of the modified Weibull distribution[J].Reliability Engineering & System Safety,2005,90(1):114-117.
[4] Ng HKT.Parameter estimation for a modified Weibull distribution[J] ,for Progressively Type-Ⅱ Censored Samples.IEEE Trans.on Reliability,2005,54 (3):374-380.
[5] 刘明治.可靠性试验[M].北京:电子工业出版社.2004;60-73.
[6] Eberhart R,Kennedy J.A new optimizer using particle swarm theory[C]//Proc of the 6th Int'l Symposium on Micro Machine and Human Science,Piscataway NJ:IEEE Service Center,1995:39-43.
[7] Wang Xiaofei.Quasi-Newton algorithm with non-monotonic trust region methods for nonlinear equations[J].Journal of Shanghai Normal University (Natural Science),2003,32 (2):16-22.
[8] 张利彪,周春光,刘小华,等.粒子群算法在求解优化问题中的应用[J].吉林大学学报(信息科学版),2005,23(4):385-389.
[9] Shi Y,Eberhart R.A modified particle swarm optimizer[C]//Proc.of the IEEE International Conference on Evolutionary Computation,Piscataway NJ,IEEE Press,1998:69-73.
[10] Shi Yuhui,Eberhart R.Parameter selection in particle swarm optimization[C]//Proc,of 7 th Annual Conference on Evolutionary Programming,1998,591-601.
[11] Clerc M.The swarm and the queen:towards a deterministic and adaptive particle swarm optimization[C]//Proc,of CEC,1999:1951-1957.
[12] The Jin Ai,Kachitvichyanukul,Voratas.Dispersion and velocity Indices for observing dynamic behavior of particle swarm optimization[C]//IEEE Congress on Evolutionary Computation[C] ,2007:3264-3271.
[13] Zhang L F,Xie M,Tang L C.A study of two estimation approaches for parameters of Weibull distribution based on WPP[J].Reliability Engineering and System Safety,2007,92(3):360-368.
[14] 戴树森,费鹤良,王玲玲,等.可靠性试验及其统计分析(下册)[M].北京:国防工业出版社,1984:7-12.
[15] 王锡清,张皓,杨广斌.恒定及步进应力加速寿命试验在电容器中的应用[J].杭州电子工业学院学报,2001,21(1):7-11.
[16] 崔应钟,吴修群.两参数威布尔分布极大似然估计软件[J].质量与可靠性,1990,2:32-33.