Hydroelectric power is widely used because it is environmental friendly, renewable and green. The cavitation is an inevitable phenomenon during the operation of hydro turbine, which is related to the efficiency and service life of the unit. This paper aims to discriminate the phenomenon of the incipient cavitation, prevent the early destruction and avoid the irreversible damage to hydro turbine.
The paper tries to find out the characteristics of cavitation entirely through a variety of features. The method comprises collection of the signals using a hydrophone, acceleration sensor and acoustic emission sensor; analyzing cavitation signal by using the way of wavelet time-frequency, peak factor and power spectral density; and comparing the different wavelet basis for analyzing signals and find the most suitable one.
The analyzed results show that the wavelet basis of morlet is more suitable for the cavitation signals. The hydrophone can distinguish the different operating conditions and discriminate the difference between the phenomenon of incipient cavitation and the other state of cavitation. The results show that when the hydrophone and acceleration sensors are used, the accuracy rate goes up to 75 per cent, which meets the requirements for the detection for incipient cavitation.
This paper focuses on finding the best sensor to discriminate the operating state of incipient cavitation to prevent early destruction.
