Imbalance fault detection based on the integrated analysis strategy for variable-speed wind turbines

With the rapid development of wind power generation, the increasing capacity of wind turbines brings the problems of high maintenance costs and high failure rate. It is very essential to detect the faults of wind turbines for enhancing safety and reliability. Because of the traditional analysis methods could not easily detect the imbalance and get clear fault characteristic frequencies, a novel analysis strategy should be discovered to resolve these issues. In this paper, the proposed fault detection analysis strategy integrates the order tracking analysis method and power spectral density method which analyzes the aerodynamic torque to detect the imbalance faults. If the faults of icing blades and angle faults occurs, the proposed analysis strategy could calculate the amplitude values of 1P and 3P frequencies in power spectrum and judge the fault types. It is more rapid, accurate and effective than the conventional fault detecting method. Meanwhile, it does not need to add additional sensors or data acquisition devices in wind power systems. The proposed analysis strategy uses order tracking analysis method to analyze the signals of input aerodynamic torque from the hub and reconstruct them. Then, the reconstructed signals could be processed by the power spectral density (PSD) analysis method to extract the fault characteristic frequencies in the order spectrum. Finally, the imbalance faults of the wind turbine could be detected easily by comparing the 1P and 3P frequencies. The model of a variable-speed wind turbine with DFIG and the integrated analysis strategy are established with G.H.Bladed and MATLAB software, respectively. The simulation results verify the proposed strategy is an effective way to detect various imbalance faults of variable-speed wind turbines.