Abstract: With the continuous expansion of offshore wind power capacity, blade lightning protection inspection—a critical safety measure—faces industry-wide challenges including inefficiency, high risks, and elevated costs. This study proposes a fully autonomous lightning protection detection system utilizing multirotor drones equipped with high-precision sensor arrays. By integrating millimeter-level positioning control, multimodal data fusion, and AI defect recognition technologies, the system achieves three key capabilities: conducting lightning arrester resistance testing, visualizing lightning damage assessment, and creating 3D structural deformation models. Verified through field tests at an offshore wind farm, the system reduces single-inspection time from 4h to 0.5h, achieving an eightfold efficiency improvement with over 95% accuracy. Compared to manual methods, it cuts costs by 40% while addressing blind spots in operations above 100m. The research also pioneers integrating China's GB/T 36295—2018 lightning protection standards with digital twin technology, establishing an integrated "inspection-evaluation-early warning" smart maintenance framework that demonstrates engineering value for digitalized offshore wind power.