Application of an improved MCKDA for fault detection of wind turbine gear based on encoder signal. (May 2020)

• Record Type: Journal Article
• Authors: Miao, Yonghao; Zhao, Ming; Liang, Kaixuan; Lin, Jing
• Is Part Of: Renewable energy. Volume 151(2020)
• Page Start: 192
• Page End: 203
• Publication Date: 2020-05
• Subjects: Wind turbines -- Rotary encoder -- Adaptive filtering -- Gearbox fault diagnosis -- Deconvolution; Renewable energy sources -- Periodicals; Power resources -- Periodicals; Énergies renouvelables -- Périodiques; Ressources énergétiques -- Périodiques; 333.794
• DOI: 10.1016/j.renene.2019.11.012
• Languages: English
• ISSNs: 0960-1481

Abstract

Abstract: Due to severe working condition, unexpected failures in wind turbine gearbox become rather frequent and may lead to long downtime or even catastrophic casualties. However, traditional diagnosis techniques based on vibration, acoustic emission etc. still face some problems when they are used for failure identification of wind turbine gearbox. Encoder signal carries rich diagnostic information which may be considered as an alternative tool for the wind turbine condition monitoring. Motivated by this, the encoder signal is initially introduced for the fault diagnosis of wind turbine gear in this paper. A novel adaptive filtering method, improved maximum correlated kurtosis deconvolution adjusted (IMCKDA), is proposed to eliminate the diverse noises in encoder signal. Additionally, to overcome the limitation from the sensibility of discontinuity point and filtered signal in traditional deconvolution methods (DMs), convolution adjustment definition is introduced. And correlated Gini index (CG) is originally designed to guide the selection of filter length. Finally, the encoder signal is verified to be an alternative tool for the fault diagnosis of wind turbine gear by real experimental cases. And without any prior knowledge and the least input parameters, IMCKDA is more suitable for processing encoder signal than existing state-of-the-art DMs. Highlights: The built-in encoder signal as a new alternative tool is firstly used for gearbox fault detection in wind turbines. An improved MCKDA is proposed to denoise the encoder signal without any prior knowledge and the least input parameters. Compared with the traditional DMs, the proposed IMCKDA is more suitable for fault diagnosis based on the encoder signal. The proposed method can expand the application to the vibration, sound and current etc. other data styles.