Vahid Pashazadeh - School of Electrical and Computer Eng., University. of Tehran
Babak N. Araabi
Farzad R. Salmasi

In this paper, a Fault Detection and Isolation (FDI) method is developed for wind turbine sensors and actuators based on a data driven approach. In the presence of high stochastic wind and noisy sensors FDI is a challenging problem. This problem is crucial when, implementation of wind turbine is very complex and high nonlinear. Thus the use of data driven method seems to be very suitable in this case. Here a Decision Tree classifier is used for Fault Detection (FD) and another Decision Tree classifier is used for Fault Isolation (FI) step. Some representative simulation results are given to show the effectiveness of the FDI method.

Wind Turbine, Fault Detection and Isolation, Feature Extraction, Wavelet Coefficient