计算机集成制造系统 ›› 2019, Vol. 25 ›› Issue (第5): 1062-1074.DOI: 10.13196/j.cims.2019.05.004

• 当期目次 • 上一篇    下一篇

面向加工时间可控的柔性作业车间节能调度问题建模

孟磊磊1,张超勇1+,肖华军1,詹欣隆1,罗敏2   

  1. 1.华中科技大学数字制造装备与技术国家重点实验室
    2.湖北汽车工业学院电气与信息工程学院
  • 出版日期:2019-05-31 发布日期:2019-05-31
  • 基金资助:
    国家重点研发计划资助项目(2016YFF0202002);国家自然科学基金面上资助项目(51575211,51875429);国家自然科学基金国际(地区)合作与交流资助项目(51861165202)。

Mathematical modeling of energy-efficient flexible job shop scheduling problem with controllable processing times

  • Online:2019-05-31 Published:2019-05-31
  • Supported by:
    Project supported by the National Key R&D Program,China(No.2016YFF0202002),the National Natural Science Foundation,China(No.51575211,51875429),and the National Natural Science Foundation for International Cooperation and Exchange,China(No.51861165202).

摘要: 针对制造车间能量利用率较低、节能潜力巨大的现状,以最小化车间总能耗和最大完工时间为目标,研究了考虑关机/重启节能策略和加工时间可控的柔性作业车间调度问题(FJSP)。首先,对考虑关机/重启节能策略和加工时间可控FJSP车间能耗进行了分析与建模;然后,根据加工时间可控FJSP特性,分别基于空闲时间与空闲能耗的建模思想,提出两个考虑关机/重启节能策略的混合整数线性规划(MILP)模型;最后,使用CPLEX求解器对20组测试实例进行求解,分别从尺寸复杂度与计算复杂度两方面对所提出的两个MILP进行对比评估。实验结果表明,所提出的两个MILP模型都是有效的,基于空闲能耗的MILP模型效果好于基于空闲时间的MILP模型。基于ε-约束法,将最大完工时间目标转换为约束条件,获得了问题的Pareto解,并进一步对所求解甘特图进行分析,挖掘了节能规则。

关键词: 柔性作业车间调度, 加工时间可控, 混合整数线性规划, 节能, &epsilon, -约束法

Abstract: Manufacturing shops are of low energy efficiency and have huge potential space for energy saving.With the minimization of workshop's total energy consumption and makespan,Flexible Job Shop scheduling Problem (FJSP) with controllable processing times and the energy-saving strategy of turning off and on was studied.Energy consumption of workshop was analyzed and modeled.Two Mixed Integer Linear Programming (MILP) models with turning off and on energy-saving strategy were formulated respectively based on two different modeling ideas,namely idle time and idle energy.The experiments of 20 instances were solved with CPLEX solver,and two MILP models were compared and evaluated not only from size complexity but also from computational complexity.The results showed that both models were effective.The model based on the modeling idea of idle energy outperformed the one based on the modeling idea of idle time.Based on ε-constraint method,Pareto front was obtained with the objective of minimizing of makespan being constraint.By analyzing the solutions from Pareto,energy-saving rules were mined.

Key words: flexible job shop scheduling problem, controllable processing times, mixed integer linear programming, energy saving, ε- constraint method

中图分类号: