摘要
货位分配问题是自动化立体仓库优化的关键。针对传统遗传算法难以收敛至全局最优解的问题,提出了一种改进遗传算法的电气设备仓库货位优化方法。该方法根据优化目标构建了数学模型,使用拉丁超立方抽样法对算法初始化环节进行优化;为了克服遗传算法的局部搜索能力差和收敛速度慢问题,使用改进自适应交叉变异及逆转操作和模拟退火操作构成改进模拟退火遗传算法。实验结果表明,相比于传统遗传算法的求解结果,改进算法显著提高了对目标函数的优化,并且其收敛性和稳定性更佳,该算法在实际工程应用中提出了有效的解决方案。
The problem of location allocation is the key to the optimization of automated three-dimensional warehouse.Aiming at the problem that the traditional genetic algorithm is difficult to converge to the global optimal solution,a location optimization method for electrical equipment warehouse based on improved genetic algorithm is proposed.A mathematical model is built with the optimization objective,and the Latin hypercube sampling method is used to optimize the initialization link of the algorithm.In order to overcome the poor local search ability and slow convergence speed of the genetic algorithm,an improved adaptive cross-mutation and reverse operation and simulated annealing operation are used.The operation constitutes a modified simulated annealing genetic algorithm.The experimental results show that compared with the solution results of the traditional genetic algorithm,the improved algorithm has significantly improves the optimization of the objective function,and its convergence and stability are better.The improved algorithm provides an effective solution in practical engineering applications.
作者
张延华
姜雄文
ZHANG Yan-hua;JIANG Xiong-wen(School of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China)
出处
《控制工程》
CSCD
北大核心
2023年第4期620-628,共9页
Control Engineering of China
关键词
自动化立体仓库
货位优化
遗传算法
初始化
自适应
Automated three-dimensional warehouse
location optimization
genetic algorithm
initialization
adaptive
