Two-Stage Scheduling for Automated Container Terminal-Microgrid Considering Logistics Flexibility

The cooperation between automated container terminals (ACTs) and microgrids (MGs) can help mitigate the operational risks associated with integrating renewable energy into MGs. However, the cooperation is complicated by the time scale mismatch between ACT minute-level task processing and autonomous guided vehicles (AGVs) scheduling (TPAS) and hourly MG operations. This paper proposes a two-stage scheduling method addressing multi-timescale ACT-MG that incorporates logistic flexibility. The first stage establishes an hourly TPAS model integrating AGV charging, transportation, and routing processes to enhance operational flexibility while ensuring feasibility. Building upon this, a multi-timescale flexibility equivalence-decomposition method is proposed, establishing flexibility boundaries via a constraint generation approach with auxiliary power limits. The second stage formulates a task-driven time-domain optimization model based on first-stage boundaries to address TPAS period coupling. The model decomposes AGV flexibility per period, reducing TPAS complexity while enabling cross-horizon coordination with MG. Additionally, to protect data privacy during ACT-MG cooperation, a customized spatial decoupling method (CSDM) is presented, which incorporates the generation of projection functions and the effective determination of initial solutions. Case studies demonstrate a 9.98% reduction in ACT-MG total cost, a 16% improvement in two-stage operational feasibility, and CSDM superior convergence (over 50% faster than traditional methods).