Fault diagnosis based on SWPT and Hilbert transform

The wide use of rolling-element bearing makes the bearing's fault diagnosis and fault evaluation of great significance. An adaptive algorithm based on stationary wavelet packet decomposition and Hilbert transform is proposed to extract the bearing fault characteristic component from the vibration signals. Then based on the extracted fault characteristic component, a fault evaluation factor is defined. Simulations show that the proposed algorithm can extract the bearing fault characteristic component with fewer computational workloads and less data storage. The change trend of the fault evaluation factor provides the suggestion of the time for the replacement of bearing. (C) 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of [CEIS 2011]