Grid Fault Detection of DFIG Wind Farms using a High-Fidelity Model and Machine Learning

A fault detection algorithm is proposed in this paper for wind farms. The algorithm is based on machine learning adopting Support Vector Machines (SVM). The focus is on the detection of grid faults for a Wind Farm of Doubly Fed Induction Generator (WF-DFIG). A high-fidelity model is used to simulate the system performance in faulty and healthy conditions. Statistical features based on steady-state signal analysis of measured signals are extracted. Machine learning results on a 9MW model WF-DFIG Grid-connected are provided by MATLAB/SIMULINK. The model performance is clarified and success in fault detection is illustrated. © 2022 IEEE.