A machine learning approach for condition monitoring of wind turbine blade using autoregressive moving average (ARMA) features through vibration signals: a comparative study. (2018)

• Record Type: Journal Article
• Authors: Joshuva, A.; Sugumaran, V.
• Is Part Of: Progress in industrial ecology. Volume 12:Number 1/2(2018)
• Page Start: 14
• Page End: 34
• Publication Date: 2018
• Subjects: condition monitoring -- wind turbine blade -- ARMA features -- machine learning -- vibration signals; Industrial ecology -- Periodicals; 658.4083
• Languages: English
• ISSNs: 1476-8917

Abstract

In wind turbine, blades are the major component for capturing the wind, however, due to environmental conditions it prompts to get a fault. To overcome this problem, a machine learning based condition monitoring technique is incorporated into the wind turbine to identify the fault classification which occurred in the blade. In this study, a three-bladed horizontal axis wind turbine was chosen and the faults like blade bend, blade cracks, hub-blade loose connection, blade erosion and pitch angle twist were considered as these are the faults mostly affect the turbine blade. In this study, the autoregressive-moving-average (ARMA) features have been extracted from the raw signal and the dominating feature was selected through J48 decision tree algorithm followed by the fault classification using machine learning classifiers. The results were compared with respect to the classification accuracy and their computational time of the classifier.