Bearing fault diagnosis based on the dimension–temporal information

This article presents a procedure for bearing fault diagnosis through incorporation of the dimension–temporal information of bearing vibration signals. By analysing bearing dynamic response to damage and the resulting vibration pattern, the procedure proposes a set of vibration features in a mutually complementary fashion. Then, the temporal information capturing the change of the bearing vibration pattern with bearing damage progression is obtained by trending analysis from a proposed vibration feature – the energy index, and combined for bearing health assessment with the dimension information of other proposed vibration features. The bearing fault progression data from motor bearing experiments validate that the procedure is effective to indicate the type and severity of bearing damage.