Research on an enhanced scale morphological-hat product filtering in incipient fault detection of rolling element bearings

Abstract

Listen

Translate article

Incipient vibration signals of rolling element bearing are usually characterized by weak fault symptoms and multiple interference source components, which imply that it is difficult to recognize effectively the defects of rolling element bearing at an early stage. To address the issue, a novel early fault detection strategy based on an enhanced scale morphological-hat product filtering (ESMHPF) is proposed in this paper. Firstly, motivated by the existing morphology theory, the concept of morphology-hat product operation (MHPO) is presented to handle the collected weak fault signal, which can extract efficiently periodic impulse characteristics closely linked to the bearing defects. Subsequently, diagonal slice spectra (DSS) are incorporated into morphological analysis, which can achieve the efficacy of noise rejection and feature enhancement. Ultimately, the optimal scale morphological filtering results are determined by using a sensitive index termed as fault feature ratio (FFR) for identifying weak damage feature and completing early fault detection. Simulated signal and two experimental cases of run-to-failure are performed to assess the efficacy of the proposed algorithm. The analysis results achieved show that the formulated algorithm can identify clearly early fault symptoms immersed in bearing vibration data. Moreover, the availability of superiority of our designed approach is demonstrated by comparing with traditional multiscale morphological filtering and some existing algorithm. This study provides a new ideafor the improvement of incipient damage identification of rolling element bearings.