DIAGNOSIS OF ROTARY MACHINES FAULTS USING ARTIFICIAL INTELLIGENCE

Abstract

This thesis employs three methods for diagnosing the bearing faults based on Fast Fourier Transform (FFT), Artificial Neural Networks (ANNs) and Adaptive Neuro-Fuzzy Inference System (ANFIS) models. The obtained data was classified into four main conditions: Healthy, Outer Faulty, Inner Faulty, and Ball Faulty. The four conditions were imported to ANN and ANFIS models. The input data was preprocessed before entering to ANN and ANFIS models by using three techniques: the normalized data in range (0-1), the time domain features, and finally the Auto Regressive (AR) model. The accomplished outcomes of ANN and ANFIS models in case of AR model give high accuracy results in classification issue.