Abstract: A task scheduling integrated optimization method based on dynamic storage allocation strategy is proposed to solve the energy consumption optimization task scheduling problem of automatic stereo warehouse with stacker. This strategy allows the reuse of empty storage locations generated by picking in the current task execution batch, resulting in task sequence constraints. Combined with the calculation difference of stacker’s running time under different motion states, a scheduling model with the goal of minimizing the total energy consumption of stacker task under compound instruction is established. According to the characteristics of the model, an improved imperialist competition algorithm is proposed to solve the model, which can satisfy the task order constraint by setting the correction mechanism, and the concept of alien population invasion is introduced to avoid the algorithm falling into “premature”. Finally, the task scheduling algorithm proposed in this paper is verified by two groups of data. The experimental results show that the proposed scheduling algorithm is superior to other methods in energy consumption and task execution efficiency, and can effectively reduce energy consumption and improve operation efficiency of stacker.

Key words: dynamic storage allocation strategy, correction mechanism, intrusion mechanism, improved imperialist competitive algorithm

摘要： 针对堆垛机式自动化立体仓库能耗优化任务调度问题，提出了一种动态储位分配策略下的任务调度集成优化方法。该策略允许在当前任务执行批次内重用拣货产生的空库位，产生了任务顺序约束；并结合不同运动状态下堆垛机的运行时间计算差异，建立了复合指令下以堆垛机任务总能耗最低为目标的调度模型。依据模型的特点，提出了一种改进帝国竞争算法进行求解。该算法通过设置校正机制使优化解能够满足任务顺序约束，并引入外来种群入侵的概念来避免算法陷入“早熟”。最后通过两组数据对提出的任务调度算法进行了验证，实验结果表明，改进算法在能耗和任务执行效率方面优于其他方法，能够有效减少堆垛机作业过程能耗，提升作业效率。

关键词: 动态储位分配策略, 校正机制, 入侵机制, 改进帝国竞争算法