10.1063/5.0062660.1

Abstract

We report for the first time a cavitation-induced pressure fluctuation decomposition developed from empirical mode decomposition (EMD) [Huang et al., Proc. R. Soc. London, Ser. A 454, 903–995 (1998)]. The idea is to decompose the nonlinear and non-stationary time series data into a finite and usually small number of “intrinsic mode functions” based on the local properties of the signal, which admit a well-behaved Fourier transform. With this transform, we can obtain frequency characteristics that give sharp identifications of imbedded structures. The cavitation evolution and excited pressure fluctuation around a cavitating propeller in the nonuniform wake are investigated using high-speed imaging and pressure sensors. By the EMD method, we separate the pressure fluctuations induced by different types of cavitation. The high frequency components of the pressure fluctuations are mainly caused by the collapse of sheet cavitation, followed by the shrinking and growth of sheet cavitation. Furthermore, the tip vortex cavitation leads to higher frequency but contributes less to pressure fluctuations. The periodical motion of the propeller contributes to the first blade frequency, and the pressure fluctuations induced by cavitation are superimposed on it.