Multi-Stage Dispatch of Seaport Power Systems for Incorporating Logistical Flexibilities in Uncertain Operational Conditions

Electrified logistical operation (LO) provides significant demand response potentials for seaport power systems. However, existing simple logistical flexibility (LF) models and attendant logistical-related uncertainties may reduce benefits and result in infeasible LO schemes. This paper proposes a multi-stage logistic-power scheduling method under variable ship arrival/ departure schedule and renewable generation. First, a comprehensive optimal logistic-power flow (OLPF) model is proposed to incorporate LF into power dispatch (PD), where both temporal and spatial effects of LO on PD are considered. Then, a multi-stage robust dynamic programming (MRDP) model is developed for the real-time dispatch of seaport power systems. LO is determined before PD when accurate ship arrival time is observed at the beginning of the current stage. The operational costs of remaining stages are minimized based on the estimated ship arrival and the worst-case renewable outputs. Following this way in rolling horizon, feasible solutions are dynamically obtained regardless of future uncertainties. The sequential decision-making process causes a nested structure of MRDP model. A mixed affine policy-based reformulation method is tailored to transform the original model into a tractable single-level mixed-integer linear programming (MILP). Case studies demonstrate the advantages of the proposed method for providing flexibility and feasibility in seaport power systems.