Data-driven Optimal Operation of Port Distribution Network Considering Logistics Load Regulation

This study presents a data-driven distributionally robust optimization method for the operation of port distribution network (PDN). Capturing the interplay between the PDN and the overall logistics system, the proposed approach integrates crucial operational aspects including berth allocation, quay crane allocation, and electric truck charging tactics. To quantify the uncertainty within a certain confidence region of renewable power forecasting error, an ambiguity set, parameterized based on historical data, is established. Building upon this ambiguity set, a distributionally robust optimization model has been developed for the operation of a PDN. Subsequently, by utilizing duality techniques, the internal maximization problem is converted into a minimization problem, which ensures the tractability of the model. Case studies are conducted on a modified 27-bus distribution network in southern China. The results verify the model's effectiveness in managing port logistics system loads, mitigating renewable power forecasting errors, and reducing PDN operational costs.