Fingerprint feature and dynamic threshold mechanism based on acoustic emission for bearing fault detection

•A condition monitoring method for visualizing the bearing state under variable-speed conditions based on fingerprint feature is established.•Threshold coefficient Rt and distribution aggregation index (DAI) are defined based on the dynamic threshold mechanism to obtain clearer fingerprint feature.•The fingerprint fault spectrum is proposed by contacting the number of fingerprint feature points per unit time with bearing fault characteristic frequency. The acoustic emission (AE) technology has emerged as a promising diagnostic tool for the damage detection of bearings. However, the conventional time-domain feature extraction based on AE hits lacks adequate consideration of the periodicity related to faults. To address this issue, taking the outer ring of a bearing as an example and studying the periodic AE hit behavior of bearings, a condition monitoring method, called fingerprint feature (FPF), is established in this study, providing a visual mode for bearing fault detection, along with the ability to track the bearing state. Since the dynamic threshold is essential to the formation of FPF, the dynamic threshold mechanism is investigated in detail, and a threshold coefficient Rt is defined to obtain the adaptive dynamic threshold, which allows instant AE hit extraction. The proposed distributed aggregation index can not only optimize Rt but also help evaluate the damage state of bearings. In addition, based on the FPF, a fingerprint fault spectrum (FFS) is proposed by associating the hit statistics per unit time with the fault characteristic frequency. Finally, the validity of the proposed method is proved by Hilbert envelope demodulation and its capability and practicality are verified using test rig data of rolling bearings and high-speed train bearings that operate close to actual working conditions. This study provides valuable insights into the development of online condition monitoring for bearings in industrial environment.