Fundamental study on vibration diagnosis for turbine generators using wavelet transform

We presented results of the fundamental study to introduce the wavelet transform to the vibration diagnosis for turbine generators. It is required to detect the vibration of turbine generators more accurately. The wavelet transform is used in many fields because it is able to visualize a phenomenon in the time-frequency domain and to detect the beginning time and the duration of it. We made a model rotor supported with two journal bearings to simulate typical vibrations as contact vibration, clearance vibration, and oil whip. The vibration phenomena were measured with displacement meters at the rotor and accelerometers at the rotor bearing, and then, visualized in the time-frequency domain by the wavelet transform. It is found that the dynamic spectra obtained by the wavelet transform of the contact vibration, the clearance vibration, and the oil whip, are different respectively. Therefore, this method is able to distinguish the typical vibration phenomena. The vibration phenomena can be detected and distinguished at the early stage.