Fault diagnosis for large-scale wind turbine rolling bearing using stress wave and wavelet analysis

Stress wave (SW) technique has been applied initially in the fault diagnosis as a new dynamic detecting method. But it is difficult to detect the fault signals due to the faintness of the fault signals and the background noise for the low speed machine. The properties of the SW and its transmission law in the large-scale wind turbine were studied in this paper. Firstly, the three-dimension contact model of the bearing was set up. The defects which occurred on the outer race and the inner race were simulated in the model. According to the model, the stress, the strain and the contact stress distribution was computed. Then, the stress and strain distribution law and the contact stress distribution law of the interface on good and fault bearing were compared. In the real-world testing, fault signals were acquired using stress wave transducer. Fault characteristic parameters were extracted and the background noise was reduced using wavelet analysis. Both simulation and real-world testing result obtained indicate that SW and wavelet transform can be the effective method in the fault diagnosis of large-scale wind turbine bearing.