Double cycling is used to improve the efficiency of quay cranes (QCs). However, higher QCs utilization increases the highest observed peak power demand of QCs, leading to a higher energy-related costs. Thus, this paper investigates the opportunities for peak shaving the power demand of QCs with double cycling by limiting the number of lifting QCs or the maximal energy demands of QCs. And some evaluation indicators are selected for assessment including the observed peak power demand, the productivity and utilization of QCs and the average waiting time of yard trucks (YTs). Considering the complexity and uncertainty of container terminals, this study establishes an agent-based simulation model to describe the operation processes of QCs, YTs and yard cranes (YCs), and visualize the power demands of QCs with double cycling. Finally, a series of computational experiments are carried out to explore the possibility of peak shaving the energy demand of QCs. The results show that both peak shaving policies are capable to reduce the power peaks of QCs with double cycling. And the policy of limiting maximal energy demands of QCs performs better which reduces the maximum energy demand as well as protecting electrical equipment without compromising higher QCs’ productivity. And the established agent-based simulation model can provide decision support for terminal operators to manage their energy consumption more smartly.

双循环用于提高码头起重机（QC）的效率。但是，更高的QC利用率会提高QC的最高峰值功率需求，从而导致与能源相关的成本更高。因此，本文研究了通过限制提升QC的数量或QC的最大能量需求，以双循环方式峰值削除QC的功率需求的机会。并选择了一些评估指标进行评估，包括观察到的峰值功率需求，QC的生产率和利用率以及堆场卡车（YT）的平均等待时间。考虑到集装箱码头的复杂性和不确定性，本研究建立了一个基于Agent的仿真模型来描述QC，YT和堆场起重机（YC）的运行过程，并通过双循环可视化QC的功率需求。最后，进行了一系列计算实验，以探索削峰降低QC能源需求的可能性。结果表明，两种削峰策略均能够通过两次循环来降低QC的功率峰值。限制QC的最大能量需求的策略执行得更好，这会降低最大能量需求并保护电气设备，而不会损害更高的QC的生产率。建立的基于代理的仿真模型可以为终端运营商提供决策支持，使其更智能地管理其能耗。限制QC的最大能量需求的策略执行得更好，这会降低最大能量需求并保护电气设备，而不会损害更高的QC的生产率。建立的基于代理的仿真模型可以为终端运营商提供决策支持，以更智能地管理其能耗。限制QC的最大能量需求的策略执行得更好，这会降低最大能量需求并保护电气设备，而不会损害更高的QC的生产率。建立的基于代理的仿真模型可以为终端运营商提供决策支持，使其更智能地管理其能耗。