Research on Fault Feature Extraction of Rotating Bearing Signal in Chemical Pump Equipment Based on SSA-VMD-HT Method

Aiming at the problem that the early fault characteristics of rotating bearings of chemical pump equipment are weak and difficult to find and realize non-contact signal extraction, an audio signal fault feature extraction method that improved Variational Mode Decomposition (VMD) based on Sparrow Search Algorithm (SSA) combined with Hilbert Transform (HT) is proposed. The SSA-VMD-HT method first optimizes VMD parameters by using Sparrow Search Algorithm, and then decomposing VMD signals to obtain a set of intrinsic mode functions (IMF), and then use correlation coefficient to select modal components containing more fault information, and finally use Hilbert Transform for instantaneous energy analysis to achieve fault feature extraction. The experimental simulation shows that SSA-VMD-HT method can reflect the characteristic information of fault signals more accurately than the traditional VMD method, and effectively improve the differentiation of fault types.