Condition monitoring techniques of the wind turbines gearbox and rotor
- Authors: Salem, Abdulwahed , Abu-Siada, Ahmed , Islam, Syed
- Date: 2014
- Type: Text , Journal article , Conference paper
- Relation: 6th International Conference on Computer and Electrical Engineering (ICCEE 2013); Paris, France; 30th-31st December 2013; published in International Journal of Electrical Energy Vol. 2, no. 1 (2014), p. 53-56
- Full Text:
- Reviewed:
- Description: Gearbox and the blades are classified as the most critical and expensive components of the wind turbine. Moreover, these parts are prone to high risk failure when compared to the rest of the wind turbine components. Due to the global significant increase in wind turbines, a reliable and cost effective condition monitoring technique is essential to maintain the availability and to improve the reliability of wind turbines. This paper aims to present a comprehensive review of the latest condition monitoring techniques for turbine gearbox and blades which are considered as the crux of any wind energy conversion system.
Optimally parameterized wavelet packet transform for machine residual life prediction
- Authors: Gondal, Iqbal , Yaqub, Muhammad , Kamruzzaman, Joarder
- Date: 2011
- Type: Text , Conference paper , Journal article
- Relation: Australian Acoustical SocietyConference 2011: Breaking New Ground, Acoustics 2011; Gold Coast, Australia; 2nd-4th November 2011; p.1-8
- Full Text:
- Reviewed:
- Description: One of the prevalent issues in condition based maintenance (CBM) is to predict the residual life of the equipment. This paper propos-es a novel framework to predict the remnant life of the equipment, called Residual life prediction based on optimally parameterized Wavelet transform and Mute-step Support vector regression (RWMS). In optimally parameterized wavelet transform, a generalized criterion is proposed to select the wavelet decomposition level which works for all the applications and decomposition nodes are selected by characterizing their dominancy level based upon relative fault signature-signal energy contents. The prediction model is based on multi-step support vector regression (MSVR) and prediction accuracy is improved in comparison with the techniques based on support vector regression (SVR). Performance of RWMS is evaluated in terms of Root Means Square Error (RMSE), studies show that proposed algorithm predicts the residual life of the equipment accurately.