Induction Motor Bearing Fault Detection based on ICA and ANN
Prashant D. Bharad, Shaila Subbaraman · 2015
Independent component analysis (ICA) is one of the robust methods to extract the features. Many researchers have indicated a great potential for this approach to analyze the signals using ICA to detect the similarity or non-similarity between two signals. We have proposed a novel method which is extension of ICA to detect the faults associated with any machine by collecting vibration signals of machine. This paper presents the details of this method. The classification of the faults, if detected, is carried out using suitable classification technique. General Terms Induction motor, ICA Algorithm, Fourier Transform, Blind Source Separation Techniques.