A method for tachometer-free and resampling-free bearing fault diagnostics under time-varying speed conditions

Abstract

Bearings often operate under time-varying rotational speed conditions. Processing the signal in the time-frequency domain and extracting the Instantaneous Fault Characteristic Frequency (IFCF) and the Instantaneous Shaft Rotational Frequency (ISRF) are important approaches for bearing fault diagnosis under time-varying speed conditions without signal resampling and without using a tachometer. However, there are two problems: (1) the collected bearing signal is often contaminated by random noise and interferences transmitted from other components, which affects the accuracy of the extracted IFCF, and (2) the ISRF cannot always be found in the Time-Frequency Representation (TFR) of the extracted bearing fault transients, which impacts the accuracy of fault identification. Therefore, a new tachometer-free and resampling-free method is proposed for bearing fault diagnosis under time-varying speed conditions which consists of three main steps: (1) bearing fault signature extraction via Oscillatory Behavior-based Signal Decomposition (OBSD) to suppress the influence of random noise and interferences, (2) IFCF and ISRF estimation via applying a IFCF&ISRF search algorithm to the TFR of the decomposed signal, and (3) automatic bearing fault identification based on the average curve-to-curve ratios of the searched IFCF and ISRF. The IFCF&ISRF search algorithm is proposed based on the analysis of the frequency characteristics of bearing vibration signals. The algorithm can estimate the ISFR even if it is not present in the extracted bearing fault signature. The effectiveness of the proposed method is validated using both simulated signals and experimental data.