Fault diagnosis of rolling bearing based on multi sensor information fusion

In order to improve the accuracy of rolling bearing fault diagnosis, this paper introduces a multi-sensor information fusion method of diagnosis. After the processing of vibration signal collected with multi-sensor, the particle swarm optimization neural networks is used for local fault diagnosis, to obtain evidence independent of each other, and then using the evidence theory fuses them. Experimental results show that the method can effectively improve the diagnostic reliability and reduce diagnostic uncertainty.