基于改进人工蜂群算法的机车车辆关键零部件可靠性研究  

Research on Reliability of Key Parts of Locomotive and Rolling Stocks Based on Optimized ABC Algorithm

作　　者：沈国强[1] 贺德强[1] 谭文举 苗剑[1] 

机构地区：[1]广西大学机械工程学院,南宁530004 [2]南宁轨道交通集团有限责任公司,南宁530025

出　　处：《中国机械工程》2018年第3期279-285,共7页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51165001);广西科技攻关项目(桂科攻1598009-6);南宁市科技攻关项目(20151021);广西制造系统与先进制造技术重点实验主任课题(15-140-30S003)

摘　　要：针对我国机车车辆现行检修制度中可能存在的过修、欠修问题,提出了一种综合图解法和改进人工蜂群(ABC)算法的机车车辆关键零部件可靠性模型拟合方法,对三参数威布尔分布进行参数估计并确立三参数威布尔可靠性模型。首先利用图解法求得三参数威布尔分布参数的初始估计值以及ABC算法的搜索空间,随后运用改进ABC算法进行迭代计算以获得较优的参数估计值,最后将可靠性模型拟合方法与最小二乘法以及概率加权矩法进行分析比较。研究结果表明,提出的可靠性模型拟合方法能够较为准确地确定三参数威布尔可靠性模型。以该模型为基础对机车车辆关键零部件进行可靠性指标计算,为决策人员确定机车车辆最佳维修周期和改进检修规程提供了科学依据。According to the over-maintenance problems and inadequate maintenance problems in the operations and maintenances of locomotive and rolling stocks based on graphical method and ABC algorithm for the models of reliability of key parts of locomotive and rolling stocks,a new fitting method was proposed to established the three-parameter Weibull model by estimating the three parameters of Weibull distribution.Firstly,the initial estimates of the three-parameter Weibull parameters and the search spaces of the ABC algorithm were determined by the graphical method.Then the ABC algorithm was used to obtain the optimal parameter estimation.Finally,a new fitting method was compared with the least squares method and method of probability weighted moments.The results show that this method may obtain the parameter estimation values and build the reliability analysis model of the key parts of the locomotive and rolling stocks accurately.The Weibull model may be used in determining the optimal maintenance periods and improve the maintenance procedures of locomotive and rolling stocks for maintenance personnels.

关 键 词：机车车辆 人工蜂群算法 三参数威布尔分布 可靠性分析 检修周期 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] U270.3[自动化与计算机技术—控制科学与工程]