基于混合粒子群算法的柴油机喷油器可靠性分布的参数估计  

DIESEL INJECTOR RELIABILITY DISTRIBUTION OF PARAMETER ESTIMATION BASED ON HYBRID PARTICLE SWARM ALGORITHM

作　　者：付涛[1] 王大镇[1] 弓清忠[1] 祁丽[1] 

机构地区：[1]集美大学机械与能源工程学院,厦门361021

出　　处：《机械强度》2014年第4期539-542,共4页Journal of Mechanical Strength

基　　金：福建省产学研重大项目资助(2012H6016);福建省自然科学基金计划资助项目(2011J01321)资助~~

摘　　要：针对在柴油机喷油器可靠性分析中运用最大似然估计求解混合威布尔参数时需要面对求解复杂联立超越方程组的问题,引入粒子群优化理论,用混合粒子群算法来求解超越方程组以达到估计混合威布尔参数的目的,并将求解结果与粒子群方法、图解法、非线性最小二乘法的求解结果进行比较。结果表明:混合粒子群算法在求解效率和收敛性能上要优于粒子群方法、图解法和非线性最小二乘法,利用该方法来估计混合威布尔分布的参数是可行的,且能获得较精确的求解结果。According to the problem that we need to solve complicated simultaneous equations when using maximum likelihood estimate to solve mixed Weibull parameters in the reliability analysis, this paper introduced the particle swarm optimization theory, used hybrid particle swarm algorithm to solve the transcendental equations to estimate the mixed Weibull parameters, compared the result with that of graphic method and nonlinear least square method. The result showed that hybrid particle swarm algorithm is better than the particle swarm method, graphical method and nonlinear least squares method in solution efficiency and convergence performance, it is feasible to estimate the mixed Weibull distribution parameters and more accurate results can be obtained using this method.

关 键 词：柴油机喷油器 混合粒子群算法 混合威布尔分布 极大似然法 参数估计 

分 类 号：TB114.3[理学—概率论与数理统计]