Open AccessSplit-Point and Attribute-Reduced Classifier Approach for Fault Diagnosis of Wind Turbine Blade through Vibration Signals
Author(s) -
A. Joshuva,
M. Arjun,
B. Sujith Adhithya,
Bilal Akash,
S. Abdul Wahaab
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/923/1/012009
Subject(s) - vibration , turbine blade , turbine , accelerometer , classifier (uml) , computer science , fault (geology) , blade pitch , statistical analysis , pattern recognition (psychology) , structural engineering , engineering , artificial intelligence , acoustics , geology , mathematics , seismology , aerospace engineering , statistics , physics , operating system
This study proposes a data processing and analysis of wind turbine blade faults using split-point and attribute-reduced classifier (SPAARC) through statistical-machine learning approach. In this study, the fault like erosion, hub-blade loose connection, pitch angle twist, bend and crack faults have been simulated and the vibration data has been taken using a piezoelectric accelerometer. With the recorded data, statistical features where extracted and with the extracted features were used to classify the fault condition on the wind turbine blade through SPAARC. The classification accuracy was found to be 85.67% and validated through 10-fold-cross-validation.