摘要
针对炼钢连铸生产调度问题具有多并行机、多缓冲、多约束的特点,提出了两阶段求解浇次序列的最小完工时间,并用启发式和遗传算法进行优化的方法。在第一阶段,以浇次为单位进行单个浇次内各炉次的分配、排序及定时,并通过有限、无限和可加工缓冲区的三类缓冲能力进行调整,获得单个浇次的初始调度;第二阶段对多个浇次调度进行合并重组,以最小完工时间为目标,运用三类缓冲区的储存能力,对各炉次的加工时间改进,优化浇次的最优序列,获得更好的调度方案。最后根据某钢厂的实例用启发式和遗传算法可获得可行性优化解。
Aiming at the characteristics of parallel-machine,multi-buffer and multi-constraint within the production of steelmaking-continuous casting,a two-phase approach which integrates heuristic rules and a genetic algorithm is proposed to minimize the makespan of casts. In the first phase,heuristic rules are adopted to determine which charges in the same cast should be scheduled on which machines in what sequence at what time. And,the scheduling plan for one single cast is further optimized through allocation of the buffers including finite capacity,infinite capacity and infinite stocking after processing. In the second phase,in order to address all casts scheduling problem with the objective of minimizing makespan,the timing of charges is further adjusted and the optimal sequence of casts is given by utilizing the storage ability of the buffers. Finally,according to the case in a steel plant,a feasible optimal solution is obtained by using the heuristic rules and a genetic algorithm.
作者
吴玲
唐秋华
张利平
韩大勇
WU Ling;TANG Qiu-hua;ZHANG Li-ping;HAN Da-yong(Key Laboratory of Metallurgical Equipment and Control Technology,Wuhan University of Science and Technology,HubeiWuhan 430081,China;Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering,Wuhan University of Science and Technology,Hubei Wuhan 430081,China)
出处
《机械设计与制造》
北大核心
2019年第3期256-260,共5页
Machinery Design & Manufacture
基金
国家自然科学基金资助项目(51275366
51305311)
关键词
炼钢连铸
两阶段
遗传算法
多缓冲
Steelmaking-Continuous Casting
Two-Phase
Genetic Algorithm
Multi-Buffer
