Multi-sensor Diagnosis Analysis of Bearing Fault Based on HOSVD Method
When different faults occur at the same time,the recognition accuracy of single channel signal is ob-viously reduced.In order to further reduce the influence of simultaneous filtering interference of bearing multi-chan-nel signal,a multi-bearing fault diagnosis method based on truncated high order singular value decomposition(HOSVD)is designed.Based on the HOSVD method,a multi-channel fault noise reduction method is proposed,and the multi-channel filter is implemented to enhance the detection efficiency to a great extent.Simulation signal analysis shows that it is difficult to extract fault feature information from time domain fault signal.The signal is ob-tained after noise reduction with the method presented in this paper.It has a good effect on noise reduction,and the final results show that it can effectively extract the periodic characteristics of the pulse.
bearingmultichannel signalfault diagnosistruncated high order singular value decompositionnoise reduction
万方数据
维普
