基于混沌自适应变异粒子群算法的铁路空车调配

摘要

It is very difficult to find an optimal solution for an actual large-scale empty car distribution problem.To solve this problem, proposed a chaos adaptive mutation particle swarm optimization algorithm.In the algorithm, enhanced the diversity of the particle swarm by using the ergodicity of the chaos to initialize the swarm, and adjusted the mutation probability by variance of the population' s fitness at each iteration, improved the capability of local and global search by introducing an adaptive inertia weighting factor for each particle to adjust its inertia weight factor adaptively in response to its fitness.Investigated the algorithm of chaos adaptive mutation particle swarm algorithm to solve railway empty car distribution problem, established the mathematic mode which minimized total distance of empty car and developed the solution algorithm.Numerical simulation results of solving railway empty car distribution problem verify that the optimum result and searching performance of chaos adaptive mutation particle swarm optimization algorithm is better than that of ACO and PSO.%为求解大规模的空车调配方案的最优解,提出了一种混沌自适应变异粒子群算法.该算法利用混沌的遍历性来初始化粒子群以增强群体的多样性,根据群体适应度方差调整变异概率的大小,并通过调整惯性权重因子以提高整个群体的全局和局部搜索能力.将该算法用于铁路空车调配,建立了以空车总走行距离最小为目标的数学模型,并在此基础上设计了相应的算法.算例结果表明该算法的寻优结果和寻优效率要优于蚁群算法和标准粒子群算法.