Multi-channel signal synchronous processing for bearing early fault diagnosis based on TRPCA

The multi-channel signal contains more complete fault information than single channel signal, which can effectively improve the fault diagnosis accuracy. Tensor decomposition based techniques are verified to have unique advantages in multi-channel signal processing. How to remove the noise interference and retain the fault characteristics information from multi-channel signal is an important problem to be solved in multi-channel synchronous diagnosis based on tensor decomposition. To effectively extract weak fault information under strong background noise, this paper studies a multi-channel signal synchronous processing method based on TRPCA for early fault diagnosis. Simulation data shows that the studied method can accurately extract weak fault characteristics in the signal under strong background noise and achieve early weak fault diagnosis.