Time-Varying Envelope Filtering for Exhibiting Space Bearing Cage Fault Features

Rolling bearings are widely used in rotating machinery and their failures may result in machine breakdown. Cage failure is of great concern because it may cause disassembly of bearing components. Since fault features of rolling bearing cage have characteristics of low frequency and small amplitude, which are often overwhelmed by strong noise and other irrelevant vibration components, there are a few research works about fault diagnosis of bearing cage, especially fault diagnosis of space bearing cage. Therefore, fault feature extraction of space bearing cage is a challenging topic. Toward this topic, a time-varying envelope filtering (TVEF) method is proposed in this article to extract the weak fault features of space bearing cage. Novel TVEF kernel functions are deduced by solving an envelope signal model, in which instantaneous frequencies (IFs) and instantaneous amplitudes (IAs) of signal components are modeled as redundant Fourier series. Fault characteristic components and their IFs and IAs can be accurately extracted by using TVEF kernel functions, which improve the signal-to-noise ratio of raw vibration signals. Based on the extracted IFs and IAs, a high-resolution time-frequency distribution is reconstructed to precisely extract the fault features of space bearing cage. Experiments designed in a flywheel test rig have verified the potential and effectiveness of the proposed TVEF method in exhibiting space bearing cage fault features than other existing methods.

Automated summary

Rolling bearing cage failure can lead to machine breakdown, with fault diagnosis of space bearing cage being a challenging topic. The proposed TVEF method accurately extracts the fault features of space bearing cage, demonstrating potential and effectiveness in experiments.