Flexibility Assessment and Hierarchical Rolling-Horizon V2G Control of EV Fleets for Photovoltaic Accommodation in Industrial Parks

The increasing penetration of photovoltaic (PV) generation in industrial parks creates challenges associated with renewable-energy accommodation due to temporal mismatches between PV output and electricity demand. Electric vehicle (EV) fleets with coordinated charging and discharging capabilities provide potential flexibility for mitigating PV curtailment and improving local energy utilization. This paper proposes a hierarchical multiobjective rolling-horizon vehicle-to-grid (V2G) control framework for EV fleets in industrial parks. First, an individual EV model is developed by considering charging/discharging modes, travel requirements, V2G willingness, and acceptable departure state-of-charge (SOC) shortfall. Based on the uncontrolled-charging baseline, pointwise fleet flexibility limits are evaluated by considering SOC constraints and departure-energy recoverability. Second, a lexicographic three-level mixed-integer linear programming (MILP) model is established to prioritize PV accommodation, followed by purchased-energy cost and user-related operating costs. The proposed optimization is implemented through a causal rolling-horizon strategy, where updated measurements and forecast information are used to revise future schedules without accessing future realized PV outputs. Case studies on a representative industrial-park system show that the causal rolling-horizon strategy increases PV utilization from 90.50% to 93.89% and reduces total operating cost by approximately 13.0% compared with the static day-ahead schedule under the tested PV deviation scenario. Furthermore, 20 paired trials with a fixed fleet composition and regenerated EV-state and PV-error realizations show consistent system-level gains, accompanied by higher user compensation and battery use. The multi-scenario comparison further shows that the proposed hierarchical rolling-horizon V2G configuration achieves 99.26% PV utilization and lower total operating cost than uncontrolled charging and unidirectional smart-charging (V1G) operation. The results indicate that the proposed framework can effectively coordinate EV-fleet flexibility and hierarchical objectives for PV accommodation in industrial parks.