Numerical simulation of cavitation erosion using coupled Euler-Lagrange models
Hydraulic and Turbo Engines and Piston Engines
Final Report Abstract
In many hydrodynamic engineering applications liquid flow vaporizes with creation of vapor cavities when entering low pressure regions. These cavities - which are usually formed by a number of vapor bubbles - may violently collapse owing to re-condensation once they reach a higher pressure region. The bubble collapse is associated to initiation of strong pressure waves. Repetitive bubble collapses in the vicinity of solid surfaces can cause significant damage by removing material from the surface. This phenomenon is known as cavitation erosion and frequently observed in maritime and hydraulic applications, e.g. ship propellers and rudders, pumps or turbines. Due to the severe risks associated to operations under sustained erosive cavitation, the accurate numerical prediction of cavitation erosion is of great importance from an industrial point of view. The present project developed a model that allows to assess the erosion risk by numerical simulations and therefore to replace model tests during the preliminary design phase. This might yield a reduction of the expenditures and time frames. Moreover, simulations are not afflicted with uncertainties due to scale effects inherent to model tests, usually performed on with a model size much smaller than the real object. The developed model tracks individual vapor bubbles as they move, grow and collapse in a liquid flow. Specific algorithms to account for bubble-bubble interaction such as collision, breakup and coalescence were introduced. The erosion assessment algorithm is based on evaluating pressure hitting the solid surface after a bubble collapse. Depending on a distance of the bubble from the surface, two models for symmetrical (away from the surface) and non-symmetrical (close to the surface) collapses are implemented. The developed approach was applied to several selected marine engineering cases such as cavitating flow over a hydrofoil and a propeller. The results display a fair predictive capability of the approach to assess erosion prone areas and flow aggressiveness. An efficient state-of-the-art parallelization strategy for the numerical algorithm was developed to run applications on supercomputers. This allows to predict erosion for challenging industrial applications with millions of vapor bubbles in a reasonable amount of wall-clock time.
Publications
- “Advanced Lagrangian Approaches to Cavitation Modelling in Marine Applications”. In: Computational Methods in Applied Sciences 29 (2013). Ed. by Luis Eca et al., pp. 217–237
Sergey Yakubov et al.
- “Experience using pressure-based CFD methods for Euler-Euler simulations of cavitating flows”. In: Computers and Fluids 111 (2015), pp. 91 –104. issn: 0045-7930
Sergey Yakubov, Thierry Maquil, and Thomas Rung
(See online at https://doi.org/10.1016/j.compfluid.2015.01.008)
