基于拥堵感知的自动化集装箱码头AGV充电策略

Congestion-aware AGV charging strategy in automated container terminal

为解决集装箱码头自动导引小车(AGV)作业过程中的拥堵、充电问题,以最小化AGV的完工时间为目标建立混合整数规划模型,提出分流拥堵路网分区中AGV的充电策略,并设计两阶段算法进行求解。第一阶段利用模拟退火算法优化AGV任务调度,第二阶段对于最优的AGV调度,进一步利用基于Dijkstra的拥堵预测算法应用拥堵感知充电策略优化AGV充电任务调度。实验表明拥堵感知充电策略比排队等待充电策略和按需充电策略分别平均节约了5.94%和2.73%的作业时间、方差分别为4.08和3.09。拥堵感知充电策略提高了AGV的作业效率,且其有效性与路段利用率、最大拥堵系数密切相关。

To solve the congestion and charging problems in the operation process of Automated Guided Vehicle(AGV)at container terminals,a mixed integer programming model was established to minimize the completion time of AGV,the charging strategy of AGV in split congestion road network was proposed,and a two-stage algorithm was designed to solve the problem.In the first stage,simulated annealing algorithm was used to optimize the AGV task scheduling.In the second stage,for the optimal AGV scheduling,the Dijkstra-based congestion prediction algorithm with congestion-aware charging strategy was further used to optimize the AGV charging task scheduling.Experimental results showed that compared with queueing charging strategy and on-demand charging strategy,congestion-aware charging strategy could save 5.94%and 2.73%of operation time,and variance were 4.08 and 3.09 respectively.The congestion-aware charging strategy improved the operational efficiency of AGV,and its effectiveness was closely related to the road utilization rate and the maximum congestion coefficient.