Mechanical Machinery Faults Detection and Classification Based on Artificial Intelligence Techniques

The bearings are vital parts for any rotating machinery and work as a key role for any machinery and dynamic system. For rotating machinery to work properly, the bearing should be healthy. This is necessary to ensure continuity and non-stop production. So, it is necessary to be supervised and monitor any rotating machinery for early detection of bearing faults. Different techniques are in use for bearing faults diagnosis such as envelope analysis, Wavelet analysis (WA), Short Time Fourier Transforms (STFT), Fuzzy Inference System (FIS) and Neural Networks (NN). This paper presents a different approach for detection and classification of bearing faults using Adaptive Neuro-Fuzzy Inference System (ANFIS). The proposed data were taken via using accelerometer which is mounted on the bearing housing in Machinery Fault Simulator (MFS). The data is measured in the X and Y directions at 2100 rev/min and 51.2 kHz rotating speed and sampling rate, respectively. Details of the design procedure and the results using the proposed method are discussed. The study is implemented offline in MATLAB environment. This article presents a comparison between the diagnosis based on ANFIS and Fast Fourier Transform (FFT) analysis. The obtained results are encouraging and promising in the field of diagnosis of mechanical machinery faults.