Impact detection for disengaged wet clutch with buckling discs using distribution distances applied to time-frequency map of vibration signal

In this paper, a new approach based on the processing of vibration signals is presented to detect impacts in disengaged wet clutches. Firstly, the STFT is applied to vibration signals to calculate a time-frequency representation. To eliminate the interference of high-energy low-frequency bands, a threshold is applied using the logarithm of variance. Then, the Bhattacharyya distance and the Hellinger distance are calculated to extract the impacts. The detection methodology is applied, tested and evaluated on nineteen acceleration vibration signals recorded on a wet clutch bench test rig. The disengaged wet clutch contains three healthy and two unhealthy discs with slight deformation. According to the relative value of the distribution distance and the time difference between each impact, the nineteen captured vibration signals are divided into three groups: periodic impacts, random impacts, and inconspicuous impacts. The proposed detection method shows high effectiveness and the preliminary analysis shows that the impacts are irregular.