基于种群多样性的自适应PSO算法求解VRPSPD问题  

Adaptive PSO Based on Swarm Diversity for VRPSPD

作　　者：罗东升[1] 刘衍民[1] 

机构地区：[1]遵义师范学院数学与计算科学学院,贵州遵义563002

出　　处：《计算机工程与科学》2012年第7期160-165,共6页Computer Engineering & Science

基　　金：贵州省科学技术基金资助项目(黔科合J字[2012]2340号);遵义师范学院科研基金资助项目(2012BSJJ19);上海市博士后基金(12R21416000)

摘　　要：为有效求解逆向物流车辆路径(VRPSPD)模型,本文提出一种基于种群多样性的自适应PSO算法(SDAPSO)。在SDAPSO运行时,根据种群多样性,自适应地对种群中运行较差的粒子进行扰动操作,提升这些粒子向最优解收敛的能力;同时,对全局最优粒子进行概率扰动,以增加种群的多样性。标准检测函数的仿真结果表明SDAPSO算法是对基本PSO算法的有效改进。在对VRPSPD模型求解中,通过与其它粒子群算法相比,表明SDAPSO是求解该类问题的一种有效方法。In order to solve effectively the vehicle routing with simultaneous delivery and pick-up problem（VRPSPD）,an adaptive PSO based on swarm diversity is proposed（SDAPSO）.In SDAPSO,the global distance disturbance is made for the worst particles in terms of swarm diversity,which improves these particles＇ ability of searching the global optimal solution.And the probability disturbance is introduced for the best performing particle（gbest） in the whole swarm to increase the diversity of swarm.In the benchmark function,the results show that SDAPSO is an effective improved algorithm compared with the basic PSO.In VRPSPD,the proposed algorithm achieves a better solution compared with other algorithms.

关 键 词：粒子群算法 种群多样性 逆向物流车辆路径问题 自适应 

分 类 号：O223[理学—运筹学与控制论]