2025 International Conference on Equipment Intelligent Operation and Maintenance (ICEIOM2025)

30 / 2025-05-08 21:21:24

GPPSE: A New Periodic Pulse Signal Extraction Method for Piston Pump Rolling Bearing

Rolling bearing fault diagnosis,Variational mode decomposition,Gini index,Blind deconvolution,Correlation analysis

The vibration patterns resulting from faults and defects in rotating machinery often exhibit distinct periodic characteristics, which serve as a critical foundation for fault diagnosis. However, the vibration signals of rotating machinery are frequently disturbed by periodic vibration noise, leading to substantial interference with the signals from the tested components and thereby compromising the efficacy of fault diagnosis methods. In this study, a comprehensive method named Gini-based Periodic Pulse Signal Extraction (GPPSE) is proposed to efficiently extract periodic pulse signals associated with rolling bearing faults in the presence of periodic noise interference. This approach can determine the optimal number of Intrinsic Mode Functions (IMF) of Variational Mode Decomposition (VMD) method based on Gini Index of Square Envelope (GISE). And it introduces a novel fault periodic pulse evaluation index, termed the G-C index, based on GISE and correlation analysis. By integrating the G-C index, VMD, and Maximum Second-order Cyclostationarity Blind Deconvolution (CYCBD), the proposed comprehensive method can effectively isolate fault periodic pulse components from rolling bearing fault signals contaminated with periodic noise. Experimental setups are designed to gather fault data from hydraulic pump rolling bearings under real-world engineering conditions, and subsequent performance verification experiments were conducted to validate the efficacy of the proposed method.