论文部分内容阅读
研究同时服务于装船和卸船作业的集卡全场调度策略,调度优化目标包括减少岸桥等待集卡的时间以及减少集卡的空载行程。提出了基于Q学习算法的集卡调度强化学习模型,对其系统状态、动作策略、报酬函数进行分析,并结合小脑模型关节控制器(CMAC)神经网络对Q函数进行泛化和逼近。仿真结果表明,与其他集卡调度策略相比,Q学习算法的优化效果比较明显,其在保证岸桥连续作业的同时,还能有效减少集卡的空载行程。
This paper studies the strategy of full-time dispatching of trucks during the process of loading and unloading operations. The objective of the dispatching optimization is to reduce the waiting time of shore quays and to reduce the no-load travel of trucks. A Q-learning algorithm based on Q learning algorithm is proposed to strengthen the learning model of card dispatching. The system states, action strategies and reward functions are analyzed. The Q function is generalized and approximated with CMAC neural network. The simulation results show that the Q learning algorithm has more obvious optimization effect compared with other set-card scheduling strategies. It can ensure the continuous quayside operation and reduce the no-load journey of the set cards effectively.