Diagnosis of Gear Faults by Using the Linear Canonical Transform in Acceleration and Deceleration Processes

Abstract

The vibration signals’ analysis is a very effective and reliable approach for diagnosing the gear faults. Since the vibration signals obtained from the gear in the acceleration and deceleration processes can often contain more beneficial fault features, the analysis of the gear vibration signals during the acceleration and deceleration processes has been a hot research topic. However, the vibration signals acquired from the gear in the acceleration and deceleration processes are nonstationary, which make the current Fourier transform (FT)-based vibration signal analysis methods invalidly. Meanwhile, the linear canonical transform (LCT) as a novel nonstationary signal analysis tool has received wide attention. Therefore, this paper proposes to use the LCT instead of the FT-based methods to diagnose the gear faults in the acceleration and deceleration processes. The simulations and the experimental evaluations are all presented to prove the validity of the LCT method in gear fault diagnosis.