摘要
针对现有研究未考虑机器多转速及优化目标不全面的问题,建立以最大完工时间、平均流经时间、总延期时长、瓶颈机器负荷、机器总负荷和系统总能耗为优化目标的多目标柔性作业车间节能调度问题(EE_HMFJSP)模型,提出一种INSGA-Ⅲ求解算法,以获得同时满足6个优化目标的调度方案。采用三段式实数编码方式进行问题表示;在工序的交叉变异阶段,设计一种概率自适应的交叉变异方法,通过控制8种交叉、变异算子的接受概率,协调个体更新机制,从而提高种群多样性;在非支配排序阶段,采用基于参考点的近似支配原则代替传统非支配原则选择个体,以降低算法的计算复杂度;最后,在测试数据集Brandimarte上进行仿真实验。结果表明:在10个算例上,INSGA-Ⅲ算法的最优值和平均值优于NSGA-Ⅲ算法和NSGA-Ⅱ算法,且该算法均取得了最小的IGD值和最大的HV值,验证了INSGA-Ⅲ算法求解EE_HMFJSP时的有效性。
To address the limitations of existing research,specifically the neglect of multi-speed machines and incomplete optimization objectives,a model was established for the multi-objective energy-efficient flexible job shop scheduling problem(EE_HMFJSP).The model incorporated six optimization objectives:makespan,mean flow time,total tardiness,bottleneck machine workload,total machine workload and total system energy consumption.Furthermore,an improved non-dominated sorting genetic algorithm Ⅲ(INSGA-Ⅲ)was proposed to generate scheduling schemes that simultaneously satisfied these six objectives.The proposed method adopted a three-segment real number encoding scheme for problem representation.During the crossover and mutation stage,a probabilistic adaptive method was designed to coordinate the individual update mechanism by controlling the acceptance probabilities of eight distinct crossover and mutation operators,thereby enhancing population diversity.In the non-dominated sorting stage,a reference-point-based approximate dominance principle was employed to replace the traditional non-dominated principle for individual selection,aiming to reduce the algorithm′s computational complexity.Finally,simulation experiments were conducted using the standard Brandimarte benchmark dataset.The results demonstrate that in 10 test cases,the INSGA-Ⅲ algorithm achieves better optimal and average values than the NSGA-Ⅲ and NSGA-Ⅱ algorithms,and it obtains the smallest IGD values and the largest HV values,verifying the effectiveness of the INSGA-Ⅲ algorithm in solving the EE_HMFJSP.
作者
栾飞
陈劲南
杨雪芹
王迪
王晓峰
LUAN Fei;CHEN Jinnan;YANG Xueqin;WANG Di;WANG Xiaofeng(College of Mechanical and Electrical Engineering,Shaanxi University of Science&Technology,Xi′an Shaanxi 710021,China;School of Electrical and Control Engineering,Shaanxi University of Science&Technology,Xi′an Shaanxi 710021,China)
出处
《机床与液压》
北大核心
2025年第23期203-209,共7页
Machine Tool & Hydraulics
基金
西安市重点产业核心技术攻关项目(23ZDCYJSGG0043-2022)
西安市科技计划项目(22GXFW0016)
西安市未央区科技计划项目(202414)。
