Abstract:

Considering the constraints between processes and machines, an optimization model with the objective of minimizing the maximum completion time or makespan was put forward to solve manifold processing shop scheduling problem effectively.To balance the ability of global and local search of the algorithm,a  MFPSO algorithm was proposed, which used staged search strategy of multi forces. The search process was divided into three stages: earlier-stage, medium-stage and later-stage, and  three kinds of force rules, were correspondingly constructed, which were single repulsion force rule, balanced attraction and repulsion force rule and single attraction force rule. Different force rules were adopted in different search stages so as to improve the search mechanism and search performance of the algorithm. The MFPSO algorithm was applied in solving manifold processing shop scheduling problem. A particle encoding and decoding method was presented based on matrix, which made use of matrix variables to deal with the constraints of the problem. Finally, the MFPSO algorithm presented herein shows better performance compared with PSO  algorithm, median-oriented PSO algorithm, extended PSO algorithm and ant colony optimization algorithm in optimizing manifold processing shop scheduling problem, thus its effectiveness was verified.

Key words: manifold processing shop scheduling, particle swarm optimization(PSO) algorithm, force rule, multi force PSO(MFPSO) algorithm

摘要：

为有效地解决液压阀块加工车间调度问题，考虑工序间和机器间的约束关系,以最大完成时间最小为目标，给出了液压阀块加工车间调度优化模型。为平衡算法的全局和局部搜索能力,提出了多作用力微粒群(MFPSO)算法，采用多作用力阶段性搜索策略，将搜索过程划分为前期、中期、后期3个阶段,并对应构造单一斥力、平衡引斥力、单一引力3种作用力规则,在不同搜索阶段采用不同的作用力规则,提高了算法的搜索机制和寻优性能。将MFPSO算法用于求解液压阀块加工车间调度问题,利用矩阵变量来处理约束条件,给出了一种基于矩阵的微粒编码、解码方法。通过液压阀块加工车间调度优化实例，将MFPSO算法与微粒群算法、中值导向微粒群算法、扩展微粒群算法、蚁群算法进行了对比,结果表明,提出的MFPSO算法结果最优，从而验证了该算法的有效性。

关键词: 液压阀块加工车间调度, 微粒群算法, 作用力规则, MFPSO算法