A New Intelligent Weak Fault Recognition Framework for Rotating Machinery

Abstract

The presence of strong background noises makes it a challenging task to detect weak fault characteristics in vibration signals collected from rotating machinery. Thus, a two-stage intelligent weak fault recognition framework, which includes signal enhancement and intelligent recognition, is proposed in this work. The signal enhancement is accomplished via an optimized relevant variational mode decomposition (ORVMD) algorithm. Specifically, the optimal parameters are derived by combining a particle swarm optimization (PSO) algorithm and the novel defined relevant energy (Re) index. This optimized VMD algorithm can extract the principal components from the raw signals. Then, the enhanced vibration signals via the proposed ORVMD are converted into spectral signals and fed into an improved stacked auto-encoder (ISAE) algorithm for fault recognition. Experimental results demonstrate the effectiveness of the proposed algorithms and fault diagnosis framework in rotating machinery fault recognition and detection.

Early intelligent fault diagnosis of rotating machinery based on IWOA-VMD and DMKELM

“…The traditional setting of K and  is based on experience or central frequency method which greatly limits the effect of signal processing. To overcome this problem, Zhao [18] optimized the VMD parameters by particle swarm optimization (PSO) with the goal of minimizing the envelope entropy and applied the optimized VMD to early fault diagnosis. Gai [19] took the minimum average envelope entropy as the objective function, used the grey wolf optimization (GWO) to adaptively obtain the optimal parameters of VMD, and combined the Teager energy operator to deal with the early fault diagnosis of bearings.…”