Title: Selection of a meta classifier-data model for classifying wind turbine blade fault conditions using histogram features and vibration signals: a data-mining study
Authors: A. Joshuva; V. Sugumaran
Addresses: Department of Mechanical Engineering, Hindustan Institute of Technology and Science, Padur, Chennai, Tamil Nadu 603103, India ' School of Mechanical and Building Sciences (SMBS), Vellore Institute of Technology, Chennai Campus, Vandalur-Kelambakkam Road, Chennai 600127, India
Abstract: The modern developments in wind turbine fault diagnosis and condition monitoring are urged in recent times. This paper aims to identify different types of faults which occur on wind turbine blade as they are prone to vibration stress due to environmental and weather condition. The fault diagnosis problem was carried out using machine learning approach. This study was carried out using vibration sources which has been acquired from good and other fault condition blades using data acquisition system. From the recorded signals, histogram features were extracted and classified using meta classifiers. From the classifiers, a better data-model is suggested for a multi-class problem in wind turbine blade fault diagnosis.
Keywords: condition monitoring; wind turbine blade; histogram features; machine learning; meta classifiers; vibration signals.
Progress in Industrial Ecology, An International Journal, 2019 Vol.13 No.3, pp.232 - 251
Received: 08 Mar 2018
Accepted: 12 Jul 2018
Published online: 19 Jul 2019 *