Assessment of cavitation erosion risk by Eulerian–Lagrangian multiscale modeling

Abstract

Cavitation erosion in fluid machineries is a complicated physicochemical and instantaneous process, leading to material loss and reduced efficiency. As the cavitation and the consequent erosion are difficult to be measured and modeled, the present work aims to incorporate an improved energy balance approach into a new Eulerian–Lagrangian multiscale model for addressing the risk of cavitation erosion. The volume of fluid (VOF) and discrete bubble model (DBM) are coupled by a transforming algorithm to develop the multiscale cavitation model. With using the large eddy simulation (LES) approach, the multiscale cavitation features including the morphological evolution of large-cavities and characteristics of microbubbles are well reproduced. To assess the cavitation erosion, a modified aggressive indicator using a modified moving time-averaged pressure field as the driving pressure is proposed to consider the evolution of local pressure and vapor volume fraction. Furthermore, the erosion powers produced from large cavities and microbubbles are completely taken into account. The results show that the VOF-DBM multiscale model can better simulate the cavitation features around a twisted hydrofoil including the secondary shedding and the horse-shoe vapor structure, compared with the VOF method using the same mesh resolution. From the results of discrete bubbles, it is found that the primary shedding leads to the rapid growth of the number of discrete bubbles, which decreases during the transportation of the bubble cloud generated from the primary shedding. By taking the collapsing power of discrete bubbles into account, the prediction of cavitation erosion distribution is quite consistent with the experimental one, whereas the VOF method fails to predict the erosion region near the trailing edge.