Fault Detection and Localization of Wind Turbine Sensors Using Output-Only Measurements

Advancements in the wind turbine technology have led to a rapid expansion in the wind energy sector placing it as one of the crucial components of the global renewable energy sectors. Wind energy farms include onshore and offshore farms. While onshore farms are cost-effective and accessible, offshore farms benefit from higher and more consistent wind speeds, although they operate in harsher conditions that increase the risk of faults. Health monitoring of wind turbines is thus essential to ensure safe operation of onshore and offshore wind turbines and to prevent any failures that can cause damages to the system and the environment. In this paper, we introduce a fault detection and localization algorithm for wind turbines sensors. The proposed algorithm, which is based on transmissibility operators, only requires measurements from sensors, which are available during the wind turbine operation. Unlike model-based techniques, the proposed algorithm does not require knowledge of a model of the wind turbine or the inputs acting on it (i.e. wind speed and direction). Moreover, unlike time-domain and frequency-domain analysis techniques, the proposed approach is not affected by changes in the operational conditions of the wind turbine. In addition, the proposed approach does not require large data sets or access to special sensors and cameras. To validate the proposed method, we utilize the FAST v8 model, a high-fidelity aeroelastic simulator developed by the National Renewable Energy Laboratory (NREL) for modeling the dynamics of wind turbines. FAST v8 provides realistic simulations of structural and aerodynamic responses, enabling a comprehensive assessment of sensor faults in various operating conditions. The proposed approach shows a false alarm rate below 1% for most sensors, and below 5% for the remaining sensors, which indicates high reliability in fault-free conditions. Moreover, the proposed approach is tested under different operating conditions of the wind turbine.