Fault diagnosis of synchronous hydraulic motor based on acoustic signals

Abstract

Synchronous hydraulic motors are used in high load conditions. Therefore, the failure of such motors must be promptly detected to avoid severe accidents and economic loss. The automation of signal processing and diagnostic processes in practical engineering applications can help improve engineering efficiency and reduce hazards. As a non-contact acquisition signal, an acoustic signal has easier acquisition than a vibration signal. This article proposes an automatic fault detection method for synchronous hydraulic motors, which uses acoustic signals. The proposed method includes the automatic calculation and pattern recognition of the parameters of fault feature vectors. The automatic calculation of the fault feature vector is based on the combination of wavelet packet energy and the Pearson correlation coefficient. Then, the nearest-neighbor classifier is used for fault diagnosis. This study verifies that the proposed method can effectively identify the normal state, gear wear, gear rust, and barrier block wear. This method provides a solution for the automatic fault diagnosis of synchronous hydraulic motors and other types of quasi-period rotating machinery.