Recent advances in structural health monitoring for offshore wind turbine blades: A review of physics-based models, data-driven methods, and surface defect detection

Due to profound complexity of marine environments, significant challenges are posed to the reliability and service life of offshore wind turbine blades (WTBs). Existing structural health monitoring (SHM) techniques that are effective for land WTBs cannot directly apply to the offshore WTBs. To comb out suitable SHM techniques for offshore WTBs, this review systematically examines the recent advances in the WTB SHM according to three interconnected pillars. First, the physics-based numerical methods for the fatigue damage assessment are introduced. Second, the data-driven fault diagnosis techniques are surveyed, elucidating their role in enhancing the detection accuracy and real-time capabilities. Third, the WTB surface defect detection methods are summarized. Lastly, a hidden connection that naturally includes these three pillars into a uniform framework is discovered by this review with the assist of the unmanned aerial vehicles (UAVs) as the sensor carrier in the marine environments; specifically, the UAV-based machine vision techniques, on one hand, can measure the WTB vibrations, and on the other hand, can perform image-based surface defect inspection using deep learning; as a result, the physics-based, data-driven and image-based methodologies are appropriated integrated into a uniform framework for offshore WTB SHM. The current deployment challenges of the UAV framework for offshore applications are discussed. Through this review, it aims to provide useful guidance and education for researchers and engineers who are planning to or engaging into the offshore WTB SHM technologies.