摘要
提出了一种新的自适应混沌粒子群优化算法。该算法在运行过程中根据群体适应度方差和最优解的大小确定当前最佳粒子引入混沌搜索有效位置的概率,有效结合粒子群全局和混沌局部搜索,避免了基本粒子群优化算法易于陷入局部最优的缺点,提高了进化后期算法的收敛精度。将该算法用于解决联盟运输调度问题,实验结果表明该算法具有较好的性能。
We present a new adaptive chaos particle swarm optimization algorithm, during the calculating process of which the probability for the current best particle to search the effective position by intoducing chaos is determined by the variance of the population＇s fitness and the current ooptimal solution. By combinning PSO with the chaotic partial searching, the proposed algorithm can void local optimum caused by basic particle swarm optimization algorithm and improve the accurancy in the later evolution period compared with original PSO. The application of this method to allied vehicle routing verifies its effectiveness
出处
《系统工程》
CSCD
北大核心
2008年第8期32-36,共5页
Systems Engineering
基金
国家自然科学基金资助项目(60374062)
广东省自然科学基金团队资助项目(8351009001000002)
广东省科技计划项目(2007B010200070)
关键词
联盟运输调度
混沌
粒子群算法
Allied Vehicle Routing Problems
Chaos
Particle Swarm Optimization
