View Item 

A sensitivity comparison of Neuro-fuzzy feature extraction methods from bearing failure signals

Access Status
Open access
Authors
Latuny, Jonny
Date
2013
Supervisor
Dr Rodney D. Entwistle
Type
Thesis
Award
PhD

Metadata
Show full item record
School
Department of Mechanical Engineering
URI
Collection
Abstract

This thesis presents an account of investigations made into building bearing fault classifiers for outer race faults (ORF), inner race faults (IRF), ball faults (BF) and no fault (NF) cases using wavelet transforms, statistical parameter features and Artificial Neuro-Fuzzy Inference Systems (ANFIS). The test results showed that the ball fault (BF) classifier successfully achieved 100% accuracy without mis-classification, while the outer race fault (ORF), inner race fault (IRF) and no fault (NF) classifiers achieved mixed results.