BEARING FAULT DETECTION USING VIRTUAL–BASED ICA ALGORITHM
K. REZAEI MOGHADAM, Amirhossein Mohajerin Ariaei, Mohammad Hossein Kahaei, Javad Poshtan · Fluctuation and Noise Letters · 2008
This paper presents a new method for fault diagnosis in ball-bearings using a combination of the Independent Component Analysis (ICA) and the Wavelet Transform. In the ICA, the number of sensors should be equal to the number of independent sources. We introduce a new method to replace the second vibration signal required by the ICA by a virtual one in order to increase the accuracy of the diagnosing system and also to simplify the system hardware. Using real and simulated signals, it is shown that the proposed algorithm outperforms the HFD algorithm.