A New Fault Diagnosis Approach Using Parameterized Time-Reassigned Multisynchrosqueezing Transform for Rolling Bearings

Rolling bearings are the core components of mechanical equipment, and their working state is closely related to the performance of mechanical equipment. When rolling bearings occur faults, the nonstationary signals will be generated. In order to effectively analyze the nonstationary signals with rapid instantaneous frequency changes, a new parametric time redistribution multisynchronous compression transform method, namely GTMSST is proposed. First, Fourier spectrum is employed to find the most suitable parametric coefficients for the kernel of General wavelet transform (GWT), and the signal is subjected to GWT to suppress the time-frequency energy diffusion. Second, a time-direction multiple iterative rearrangement strategy is developed to solve the problem that the time-reassigned synchrosqueezing is only applicable to the weak frequency signals and easily interfered with the noise, which can achieve higher time-frequency resolution and energy aggregation. Finally, the effectiveness of the GTMSST in analyzing nonstationary signals is proved through the experiments with numerical signals and actual fault signals. The results show that the GTMSST has high time-frequency energy concentration, and takes on better performance under different noise levels and operating working conditions. It can effectively diagnose the bearing faults.