A Coordinated Optimization Method for Peak-Shaving Operation of Key Port Crane Equipment

Intelligent and green development is an inevitable trend for future port development. After electrification upgrades, key port equipment is supplied by the port microgrid, and a safe and stable power supply is the foundation for efficient port transportation operations. Focusing on critical large-scale port cranes in automated container terminals, this study investigates a coordinated operational scheduling strategy for peak shaving and valley filling in port microgrids. A unified load-calculation model for ship-to-shore cranes (STS) and rail-mounted gantry cranes (RMG) is established. By dividing the stages of the equipment's dual-cycle operating process, a multi-stage, full-time-horizon load profile under standard working conditions is developed. Building on the load modeling and considering the charge-discharge capability of centralized port energy storage, an exact optimization model is proposed for the coor-dinated operation between the operating cycles of crane clusters and the energy storage system. Logical variables are introduced to characterize the operating stage and power level of each machine, and constraints and an objective function are formulated to account for aggregated power impacts from equipment clusters. Solving the optimization model yields the optimal start times for multiple machines, thereby smoothing load fluctuations of port equipment clusters and improving energy utilization efficiency. Simulation analyses are conducted for equipment operation and energy storage capacity configuration, verifying that the multi-equipment coordinated scheduling method can effectively reduce the peak-valley difference, achieving more than a 20% reduction in peak power in the considered scenarios. Comparative analyses between exact optimization and heuristic optimization validate the optimality and scalability of the exact approach. In addition, the feasible region of start times under stochastic equipment operation is provided, identifying safe start-time combinations under non-ideal conditions. An extension of the proposed optimization model to non-standard operating conditions is also presented, demonstrating strong engineering practicality.