A novel denoising method of the hydro-turbine runner for fault signal based on WT-EEMD

Sediment wear is a significant cause of hydro-turbine failure. The wavelet threshold and ensemble empirical mode decomposition (WT-EEMD) method is proposed to denoising the acoustic vibration signals of hydroturbine runners under normal and sand-laden water flow conditions. Ensemble empirical mode decomposition (EEMD) is performed on the acquired signals, and the decomposed high-frequency intrinsic mode function (IMF) is denoised using wavelet threshold. A nonlinear threshold function is constructed instead of the traditional threshold function in the wavelet threshold algorithm. The experimental results show that the WT-EEMD method is superior to the EMD, EEMD, and wavelet threshold. Moreover, it was found that when the sand-laden water flows through the hydro-turbine, it causes a change in the frequency spectrum. This study can provide a reference for the study of sand avoidance operation of hydro-turbines and provide a valuable supplement to the existing condition monitoring and fault diagnosis system of hydroelectric generators.