Wind turbine gearbox condition monitoring based on class of support vector regression models and residual analysis

Dhiman, Harsh S. and Deb, Dipankar and Carroll, James and Muresan, Vlad and Unguresan, Mihaela-Ligia (2020) Wind turbine gearbox condition monitoring based on class of support vector regression models and residual analysis. Sensors, 20 (23). 6742. ISSN 1424-8220 (https://doi.org/10.3390/s20236742)

[thumbnail of Dhiman-etal-Sensors-2020-Wind-turbine-gearbox-condition-monitoring]
Preview
 Text. Filename: Dhiman_etal_Sensors_2020_Wind_turbine_gearbox_condition_monitoring.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (735kB)| Preview

Abstract

The intelligent condition monitoring of wind turbines reduces their downtime and increases reliability. In this manuscript, a feature selection-based methodology that essentially works on regression models is used for identifying faulty scenarios. Supervisory control and data acquisition (SCADA) data with 1009 samples from one year and one month before failure are considered. Gearbox oil and bearing temperatures are treated as target variables with all the other variables used for the prediction model. Neighborhood component analysis (NCA) as a feature selection technique is employed to select the best features and prediction performance for several machine learning regression models is assessed. The results reveal that twin support vector regression (99.91%) and decision trees (98.74%) yield the highest accuracy for gearbox oil and bearing temperatures respectively. It is observed that NCA increases the accuracy and thus reliability of the condition monitoring system. Furthermore, the residuals from the class of support vector regression (SVR) models are tested from a statistical point of view. Diebold–Mariano and Durbin–Watson tests are carried out to establish the robustness of the tested models.

CORE (COnnecting REpositories)