Incipient Fault Feature Extraction of Rolling Bearing Based on Signal Reconstruction

• DOI: 10.3390/electronics12183749
• Publication Date: 2023-09-05T13:59:31Z
• Authors: Xu Lv; Fengxing Zhou; Bin Li; Baokang Yan
• Keywords: SIGNAL (programming language); Signal reconstruction

In the incipient fault vibration signals of rolling bearings, weak features are easily submerged in strong background noise and difficult to be extracted. The sparse decomposition method can perform well extraction features, but low signal-to-noise ratio (SNR) would cause excessive decomposition. To enhance maintain time–frequency structure impulses, a novel feature bearing based on signal reconstruction is proposed. Firstly, Teager energy operator (TEO) used obtain envelope impulse components signal, which also sensitive seriously affected by noise. Secondly, Savitzky–Golay (S–Golay) filter particle swarm optimization (PSO) algorithm adopted suppress TEO generate smooth signal. Finally, reconstructed multiplication filtered original structural characteristics impact provide reliable enhancement for further decomposition, multi-source separation, other operations. Simulation experiments verify effectiveness this extracting early under SNRs.