Fault Diagnosis and Analysis of Industrial Robot Multi-Joint Bearing Based on HOSVD
Industrial robots bear high-frequency load movements in joint bearings during operation,and single-channel signal identification cannot guarantee fault identification accuracy.In order to further reduce the interference of multi-channel signals of bearings,a multi-bearing fault diagnosis method based on truncated high-order singular value decomposition(HOSVD)for industrial robots is designed.The results show that compared with the original signal,the vibration signal noise is completely removed after applying HOSVD processing,and the pulse characteristics are effectively retained,which verifies that the design method in this paper is more reasonable.Compared with the single-bearing fault diagnosis results,multiple bearings have obvious advantages,and the diagnostic accuracy is more than 99％,which fully proves the feasibility of the multiple-bearing fusion model.This research helps to improve the service life of industrial robots and plays a good role in energy saving.
industrial robot bearingsmulti-channel signalsfault diagnosistruncated higher-order singular value decomposition
NETL
NSTL
万方数据
