Custom Simplified Machine Learning Algorithms for Fault Diagnosis in Electrical Machines

With advancements in science, machine learning and artificial intelligence integration with different fields have opened up new horizons. In this paper, some simplified custom machine learning algorithms are defined to train different faults for electrical machines. The industry has been moving towa...

Full description

Saved in:
Bibliographic Details
Main Authors: Raja, Hadi Ashraf, Asad, Bilal, Vaimann, Toomas, Kallaste, Ants, Rassolkin, Anton, Belahcen, Anouar
Format: Conference Proceeding
Language:English
Subjects:
artificial intelligence
Electrical engineering
Fault diagnosis
Industries
Machine learning
Machine learning algorithms
neural network
Neural networks
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With advancements in science, machine learning and artificial intelligence integration with different fields have opened up new horizons. In this paper, some simplified custom machine learning algorithms are defined to train different faults for electrical machines. The industry has been moving towards predictive maintenance of machines rather than scheduled maintenance with the new industry 4.0 revolution. It has also paved the way for researchers to explore more in machine learning and have specific machine learning training algorithms catered to diagnose faults in electrical machines. Here, three different variations of a simplified machine learning algorithm are present for the training of faults of electrical machines. A comparison of the results is presented at the end, along with further studies carried out in this area.
ISSN:2464-708X
DOI:10.1109/Diagnostika55131.2022.9905174