Rolling Bearing Fault Diagnosis Using Improved Deep Residual Shrinkage Networks

To improve feature learning ability and accurately diagnose the faults of rolling bearings under a strong background noise environment, we present new shrinkage function named leaky thresholding to replace soft in deep residual networks (DRSNs). In this work, discover that such improved (IDRSNs) can be realized by using group searching method optimize slope value thresholding, IDRSNs more effectively eliminate signal features. We highlight our techniques significantly performance on various fundamental tasks. Experimental results show achieve better fault diagnosis noised vibration signals compared with DRSNs. Moreover, also provide normalized processing further diagnosing accuracy bearing environment.