Collaborative Opportunistic Maintenance Scheduling for Offshore Wind Farms Considering Cognitive Correction

High operation and maintenance (O&M) costs and harsh marine environments pose significant challenges to offshore wind farms (OWFs). Existing maintenance strategies often rely on imperfect state information, leading to sub-optimal scheduling. To address these issues, this paper proposes a dynamic collaborative O&M scheduling framework utilizing a heterogeneous fleet of Crew Transfer Vessels (CTVs) and Unmanned Aerial Vehicles (UAVs). First, a reliability model based on the Weibull distribution is utilized to characterize the degradation process of wind turbines. Second, a cognitive correction mechanism is introduced, where UAV inspections dynamically update belief states and trigger event-driven re-planning when cognitive bias exceeds a safety threshold. Finally, a rolling horizon optimization model based on Deferral Opportunity Cost is established to balance deterministic dispatching costs against probabilistic future risks. Simulation results on a 40-turbine OWF demonstrate that the proposed method reduces total O&M costs significantly compared to blind scheduling and reactive strategies, effectively minimizing downtime while ensuring high reliability.