钢铁合同计划及其PSO-GA混合算法的研究  

The Research on the Order Planning of Steel Production and its PSO-GA Hybrid Algorithm

作　　者：杨乐[1] 蒋国璋[1] 刘清雄[1] YANG Le;JIANG Guo-zhang;LIU Qing-xiong(College of Machinery and Automation,Wuhan University of Science and Technology,Hubei Wuhan 430081,China)

机构地区：[1]武汉科技大学机械自动化学院

出　　处：《机械设计与制造》2019年第6期13-16,共4页Machinery Design & Manufacture

基　　金：钢铁生产混合流程智能调度及其知识网系统的研究(71271160)

摘　　要：通过对钢铁一体化生产流程的分析,建立了合同计划的多目标自然数模型,考虑到交货时间窗提前/拖后期的惩罚、设备产能约束、合同取消惩罚以及生产费用,提出通过种群分割策略以及基于粒子群算法重构遗传算法的变异算子,形成粒子群、遗传算法混合算法对模型求解,使得生产周期最短、取消合同数量最少、热装比最大以及生产成本最小。通过实例证明,粒子群、遗传算法混合算法比经典遗传算法在解决多批次、小批量订单合同计划问题方面,收敛更早,算法性能更加稳定。It has established a natural Number model of the multi-objective contract project by analyzing the process of the integrated iron and steel production.The earliness-tardiness penalty,later delivery penalty in delivery time window,equipment capacity constraint,contract cancellation penalty as well as production cost have been taken into account to get the objectives that are the shortest production cycle,the minimum number of the contract which would be canceled,the maximum of the hot charge-ratio as well as the minimum production cost.At last,It swarm-genetic and genetic algorithm hybrid algorithm which is formed by reconstructing population segmentation strategy and mutation operator of genetic algorithm based on particle swarm optimization is put forward to solve the model.The experimental results show,the hybrid algorithm converges more earlierly and performs more stablely than classical genetic algorithm to solve large batches and small quantities order planning problem.

关 键 词：钢铁一体化生产 合同计划 粒子群算法 遗传算法 

分 类 号：TH16[机械工程—机械制造及自动化] TP182[自动化与计算机技术—控制理论与控制工程]