AUTHOR=Arockia Dhanraj Joshuva , Alkhawaldeh Rami S. , Van De Pham , Sugumaran V. , Ali Najabat , Lakshmaiya Natrayan , Chaurasiya Prem Kumar , S. Priyadharsini , Velmurugan Karthikeyan , Chowdhury Md Shahariar , Channumsin Sittiporn , Sreesawet Suwat , Fayaz H. 

TITLE=Appraising machine learning classifiers for discriminating rotor condition in 50W–12V operational wind turbine for maximizing wind energy production through feature extraction and selection process

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.925980

DOI=10.3389/fenrg.2022.925980

ISSN=2296-598X

ABSTRACT=Wind power is one of nature’s important green energy assets, and one of the major renewable power resources due to its reliability. Using wind turbine blades, wind power is turned into electric power. The wind turbine blade size varies from 25 m to 120 m according to needs and efficiency required. The blades are vulnerable to different friction forces due to ambient factors and wide-structure, which can affect the blades. This results in the development of electricity and the shutdown of turbines being liable. According to structural health management, downtimes are shortened when the blades are diagnosed continuously. This paper examines the prediction of damage by means of vibration signals on the 50W-12V wind turbine. The machine learning (ML) approach creates a non-linear relation among chosen critical damage characteristics as well as the associated uniqueness measures. Based on the good state of the edge, the learning algorithm was trained and tested. Classifier models such as Naive Bayes (NB), Multi-Layer Perceptron (MLP), Linear SVM (linear_SVM), One- Deep Convolutional Neural Network (1DCNN), Bagging, Random forest (RF), XGBoosts, and Decision Tree J48 (DT) were used in this analysis to forecast blade faults and the results of the assessment were compared according to their parameters to propose a better model of fault diagnostics.