Current Sensor Fault Diagnosis in DFIG Wind Turbines: Kalman Filter-Based Dedicated Observer Scheme
Mohammed Abbas, Sid Ahmed El Mehdi Ardjoun, Houcine Chafouk · 2024
In wind farms, the accuracy of the measurements provided by current sensors is crucial to ensure effective monitoring and control precision. Malfunctioning of these sensors can lead to severe damages, particularly at the converter, generator, and network levels. Therefore, it is essential to quickly identify faulty sensors to ensure the proper functioning of the system, to avoid energy losses, and to reduce maintenance costs. This article presents a diagnostic method based on an array of Extended Kalman Filters (EKF), developed following the Dedicated Observer Scheme (DOS), relying on a nonlinear model of the Doubly Fed Induction Generator (DFIG). This approach aims to estimate the rotor currents, which are then compared with sensor measurements to produce residuals used as fault indicators to detect faulty sensors. This method has been implemented on an FPGA board, and the experimental results obtained demonstrate its effectiveness, even under varying wind conditions.