Motor bearing fault diagnosis using pattern recognition machine learning technique

摘要

This research article reveals how motor bearing faults can be diagnosed while implementing artificial intelligence. Various faults were introduced in the bearings components and their vibrations analyzed. For identification of faults, pattern recognition neural networks were used. The gathered vibration data were classified into 3 classes and then fed to the neural network model to train. A neural network model using one hidden layer with 40 neurons and conjugate gradient training algorithm was used in this research. This model can successfully classify the vibration data with 78.5 % accuracy.