Application of the Hilbert-Huang Transform with Fractal Feature Enhancement for Failure Types Classification of Wind Turbines

Abstract

A methodology for common failure types classification of wind turbine based on the Hilbert-Huang Transform (HHT) with Fractal feature enhancement is proposed. Firstly, failure types of wind turbines were established according to the frequency of happened for wind turbine. The current of generators from pre-failure wind turbines under operating are measured for every failure type. Secondly, the HHT is applied on each measured current signals and converted them into 3D energy spectrum that can reveal the time, frequency and energy distribution of signal. Then, using fractal theory to extract the useful pattern features from HHT 3D energy spectrum and also to reduce the scale of input data for classification. Finally, through observe these features, it could recognize different physical characteristic of each failure type. Therefore, subtractive clustering method for failure types classification is developed for wind turbine failure detection. To demonstrate the effectiveness of the proposed approach, comparative studies had been conducted on 120 sets of field-tested patterns of wind turbines. The results show that the proposed approach that using HHT analysis and combined with fractal theory to extract failure features can effectively identify which failure types the wind turbine belongs and also reduce classification time efficiently.