一种基于多种群NSGA-Ⅱ的卷烟厂多AGV调度方法  

NSGA-Ⅱ Based on Multi-Population for Multi-AGV Scheduling in Cigarette Factory

作　　者：李勇健 吴珂雨 

机构地区：[1]江苏中烟工业有限责任公司南京卷烟厂,江苏南京210019

出　　处：《工业控制计算机》2024年第12期122-123,139,共3页Industrial Control Computer

摘　　要：在卷烟厂生产实践中,自动导引车是助力生产自动化和智能化的重要运输工具。针对AGV应用过程中出现的AGV数量和系统作业效率的冲突,以用车数量最小化和系统作业时间最小化为优化目标,构建卷烟厂多AGV调度优化模型。为求解该问题,设计一种多种群的NSGA-Ⅱ算法,引入多种交叉算子强化算法的全局探索能力,构建三个群从不同方向对Pareto最优解进行探索,设计启发性邻域搜索策略提高算法收敛性。基于实际工作环境,构建了四个规模不同的测试算例,并通过与其他经典的多目标优化算法对比,验证所提算法能够获取收敛性和分布性较好的Pareto解集,为AGV调度提供决策方案。Aiming at the contradictory between the number of AGVs and the operating efficiency during the process of AGV application,the multi-AGV scheduling optimization model for the ingredient transportation system of a cigarette factory is established,to minimize the vehicle number and the makesapn.To solve this problem,a multi-population NSGA-Ⅱ algorithm is designed,where a variety of crossover operators are introduced to strengthen the global exploration ability of the algorithm,three subpopulations are constructed to explore Pareto optimal solutions from different directions,and the heuristic neighborhood search strategy is designed.According to the practical working environment,four test instances with different scales are constructed.By comparing with other classical multi-objective optimization algorithms,it is verified that the proposed algorithm can obtain the Pareto solution set with good convergence and distribution,providing decision schemes for AGV scheduling.

关 键 词：卷烟厂 多种群 NSGA-Ⅱ 多AGV调度 变邻域搜索 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置] TP18[自动化与计算机技术—控制科学与工程] TS48[农业科学—烟草工业]