基于混合PSO-ACO算法的液压系统可靠性优化  

Reliability Optimization of Hydraulic System Based on Hybrid PSO-ACO Algorithm

作　　者：陈东宁[1,2] 张瑞星[1,2] 姚成玉[3] 

机构地区：[1]燕山大学河北省重型机械流体动力传输与控制实验室,河北秦皇岛066004 [2]先进锻压成形技术与科学教育部重点实验室(燕山大学),河北秦皇岛066004 [3]燕山大学河北省工业计算机控制工程重点实验室,河北秦皇岛066004

出　　处：《机床与液压》2013年第23期157-161,共5页Machine Tool & Hydraulics

基　　金：河北省自然科学基金资助项目(E2012203015);河北省教育厅资助科研项目(ZH2012062);秦皇岛市科技支撑计划项目(2012021A078)

摘　　要：为降低构造复杂系统可靠性优化模型的难度,利用T-S故障树构造系统故障率函数,并结合可靠性费用函数构造可靠性优化模型。针对PSO算法局部收敛性差、ACO算法搜索初期积累信息素占用时间较长的不足,将PSO算法和ACO算法混合,并结合死亡罚函数法构造适应度函数,提出混合PSO-ACO算法。考虑不同的粒子个数和蚂蚁个数,将所提算法应用于液压工作系统的可靠性优化,通过与PSO算法、ACO算法及ACO-PSO算法的对比,验证混合PSO-ACO算法的优化结果更为理想。To reduce the difficulty of constructing the reliability optimization model of complex system, the T-S fault tree was ap- plied to compose the system failure rate function, and then the system reliability optimization model was constituted combining with the reliability cost function. To solve the lack of the local convergence in the PSO algorithm and low efficiency in pheromone accumulation at the beginning of search in the ACO algorithm, the PSO algorithm and the ACO algorithm were integrated, and the fitness function was formed combining with death penalty function method, hence the hybrid PSO-ACO algorithm was proposed. The optimal results of the hybrid PSO-ACO algorithm are better than the PSO algorithm, the ACO algorithm and the ACO-PSO algorithm, which are verified by reliability optimization examples of the hydraulic work system in different particle number and ant number cases.

关 键 词：液压系统 可靠性优化 T—S故障树 混合PSO-ACO算法 

分 类 号：TB114.3[理学—概率论与数理统计]