Chaotic eye-based fault forecasting method for wind power systems

This study proposes a method for detecting possible faults in wind turbine systems in advance such that the operating state of the fan can be changed or appropriate maintenance steps taken. In the proposed method, a chaotic synchronisation detection method is used to transform the vibration signal into a chaos error distribution diagram. The centroid (chaotic eye) of this diagram is then taken as the characteristic for fault diagnosis purposes. Finally, a grey prediction model is used to predict the trajectory of the feature changes, and an extension theory pattern recognition technique is applied to diagnose the fault. Notably, the use of the chaotic eye as the fault diagnosis characteristic reduces the number of extracted features required, and therefore greatly reduces both the computation time and the hardware implementation cost. From the experimental results, it is shown that the fault diagnosis rate of the proposed method exceeds 98%. Moreover, it is shown that for oil leaks in the gear accelerator system, the proposed method achieves a detection accuracy of 90%, whereas the multilayer neural network method achieves a maximum accuracy of just 80%.