Open AccessFault diagnosis of wind turbine planetary gearbox based on order analysis and divergence index
Author(s) -
Chengbing He,
Shunkai Cai,
Dakang Sun,
Lei Song
Publication year - 2017
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2017.0560
Subject(s) - divergence (linguistics) , fault (geology) , turbine , signal (programming language) , control theory (sociology) , hilbert–huang transform , wind power , time domain , feature (linguistics) , computer science , frequency domain , resampling , algorithm , mathematics , engineering , artificial intelligence , statistics , energy (signal processing) , geology , aerospace engineering , computer vision , philosophy , linguistics , control (management) , electrical engineering , seismology , programming language
Gearbox is one of the key components of wind turbine, whose fault will directly affect the safety and the operation of the overall wind turbine. This study develops a fault diagnosis method of wind turbine planetary gearbox. Using the order resampling method, the original non‐stationary time‐domain vibration signal is converted into a signal having a smooth angular domain or quasi‐stationary characteristics. The angle domain signal is conducted by empirical mode decomposition method. According to correlation coefficient criteria, the information mode function that contains fault feature is selected for signal reconstruction. The reconstructed signal is carried on by spectrum analysis. According to the classification of planetary gearbox, the fault feature parameters are extracted. J‐divergence and KL‐divergence (Kullback–Leibler divergence) calculated through the fault feature parameters are used to determine the fault location and describe the severity degree of the fault. Results of the experiments have revealed that the proposed method can be effective and accurate to the fault diagnosis of wind turbine planetary gearbox.