改进的Q-learning蜂群算法求解置换流水车间调度问题  

Improved Q-learning Bee Colony Algorithm to Solve the Scheduling Problem ofthe Permutation Flow Shop

作　　者：杜利珍[1] 宣自风 唐家琦 王鑫涛 DU Lizhen;XUAN Zifeng;TANG Jiaqi;WANG Xintao(Hubei Digital Textile Equipment Key Laboratory,Wuhan Textile University,Wuhan 430200,China)

机构地区：[1]武汉纺织大学湖北省数字化纺织装备重点实验室,武汉430200

出　　处：《组合机床与自动化加工技术》2024年第10期175-180,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家重点研发计划项目(2019YFB1706300)。

摘　　要：针对置换流水车间调度问题,提出了一种基于改进的Q-learning算法的人工蜂群算法。该算法设计了一种改进的奖励函数作为人工蜂群算法的环境,根据奖励函数的优劣来判断下一代种群的寻优策略,并通过Q-learning智能选择人工蜂群算法的蜜源的更新维度数大小,根据选择的维度数大小对编码进行更新,提高了收敛速度和精度,最后使用不同规模的置换流水车间调度问题的实例来验证所提算法的性能,通过对标准实例的计算与其它算法对比,证明该算法的准确性。For the scheduling problem in permutation flow shop,an artificial bee colony algorithm based on an improved Q-learning algorithm is proposed.This algorithm designs an improved reward function as the environment for the artificial bee colony algorithm.The quality of the reward function is used to determine the optimization strategy for the next generation population.Through Q-learning,intelligent selection of the dimensionality size for updating the artificial bee colony algorithm′s food sources is achieved.The selected dimensionality size is used to update the encoding,thereby improving the convergence speed and accuracy.Finally,instances of permutation flow shop scheduling problems of different scales are used to validate the performance of the proposed algorithm.Through computation on standard instances and comparison with other algorithms,the accuracy of the algorithm is demonstrated.

关 键 词：Q-learning算法 人工蜂群算法 置换流水车间调度 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]