混合分解多目标进化算法求解绿色置换流水车间调度问题  

A hybrid multi-objective evolutionary algorithm based on decompositionfor green permutation flow-shop scheduling problem

作　　者：罗聪 龚文引[1] LUO Cong;GONG Wen-yin(School of Computer Science,China University of Geosciences,Wuhan 430074,China)

机构地区：[1]中国地质大学计算机学院,武汉430074

出　　处：《控制与决策》2024年第8期2737-2745,共9页Control and Decision

基　　金：国家自然科学基金项目(62076225);湖北省自然科学基金项目(2019CFA081).

摘　　要：针对考虑能量消耗的绿色置换流水车间调度问题,以最大完工时间和总能量消耗为优化目标,提出一种混合分解多目标进化算法(HMOEA/D).首先,为了保持初始种群的多样性,使用一种混合初始化策略产生高质量初始种群;其次,采用禁忌搜索策略作为局部搜索算子,强化算法跳出局部最优能力;最后,提出节能策略,以进一步优化总能量消耗目标.通过对标准测试集进行仿真实验并与代表性算法进行比较,验证所提出算法的优越性.For the green permutation flow-shop scheduling problem(GPFSP)with energy consumption,a hybrid multiobjective evolutionary algorithm based on decomposition(HMOEA/D)is proposed,which aims at optimizing both of the makespan and total energy consumption.Firstly,in order to maintain the diversity of the initial population,a hybrid initialization strategy is used to initialize the population.Then,the tabu search strategy is used as local search operator to make the population jump out of local optima.Finally,an energy-saving strategy is proposed to further optimize the total energy consumption objective.Through the simulation of the standard benchmarks,and compared with the representative algorithm,the experimental results show that the proposed algorithm achieves better performance.

关 键 词：置换流水车间调度 绿色调度 禁忌搜索策略 节能策略 分解多目标进化算法 多目标优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]