Numerical simulation and analysis of multi-scale cavitating flows using a hybrid mixture-bubble model
Doctoral thesis, 2020
The aim of this research is to model and analyse multi-scale cavitating flows with a certain emphasis on small sub-grid vapour structures. Cavitating flows include vapour structures with different length scales, from micro-bubbles to large cavities. The correct estimation of small-scale cavities can be as important as that of large-scale structures, since cavitation inception as well as the resulting noise, erosion, pressure shocks and strong vibrations occur at small time and length scales. For numerical analysis, while popular homogeneous mixture models are practical options for simulation of large-scale flows, they are normally limited in representation of the small-scale cavities due to high computational expenses and inherent simplifications. In this study, a hybrid cavitation model is developed by coupling a homogeneous mixture model with a Lagrangian bubble model. In this model, large cavity structures are modelled using a mixture model, while small sub-grid structures are tracked as Lagrangian bubbles.
The coupling of the mixture and the bubble models is based on an improved algorithm which is compatible with the flow physics and the governing equations are revised to take into account the bubble effect on the continuum flow.
The Lagrangian bubble model is based on a four-way coupling approach in which various effective forces on bubble transport are taken into account and a new algorithm is introduced to model bubble-bubble collisions. Besides, the bubble dynamics is calculated based on the local pressure effect by introducing an improved form of the Rayleigh-Plesset equation. The other contributions include implementing a new submodel for prediction of bubble break-up as well as correcting the bubble wall boundary condition and revising the void handling scheme.
Apart from the model development, for validation of the solver, a set of experimental tests on cavitating flow around a surface-mounted bluff body are performed in this study. Then, a multi-scale test case is simulated using both the new hybrid model and the traditional mixture model. The comparison of the results with the experimental data shows considerable improvements in both predicting the large cavities as well as capturing the small-scale structures using the hybrid model. More accurate results (as compared to the traditional mixture model) can be achieved even with considerably lower mesh resolution. The results, among others, show that small-scale cavities not only are important at the inception and collapse steps, but also influence the development of large-scale structures.
The coupling of the mixture and the bubble models is based on an improved algorithm which is compatible with the flow physics and the governing equations are revised to take into account the bubble effect on the continuum flow.
The Lagrangian bubble model is based on a four-way coupling approach in which various effective forces on bubble transport are taken into account and a new algorithm is introduced to model bubble-bubble collisions. Besides, the bubble dynamics is calculated based on the local pressure effect by introducing an improved form of the Rayleigh-Plesset equation. The other contributions include implementing a new submodel for prediction of bubble break-up as well as correcting the bubble wall boundary condition and revising the void handling scheme.
Apart from the model development, for validation of the solver, a set of experimental tests on cavitating flow around a surface-mounted bluff body are performed in this study. Then, a multi-scale test case is simulated using both the new hybrid model and the traditional mixture model. The comparison of the results with the experimental data shows considerable improvements in both predicting the large cavities as well as capturing the small-scale structures using the hybrid model. More accurate results (as compared to the traditional mixture model) can be achieved even with considerably lower mesh resolution. The results, among others, show that small-scale cavities not only are important at the inception and collapse steps, but also influence the development of large-scale structures.
Homogeneous mixture model
Lagrangian bubble model
CFD
Cavitation
OpenFOAM
Multiphase flow
Hybrid model
Multi-scale
Lecture hall HA1, Hörsalsvägen 4, Chalmers
Opponent: Prof. Dr.-Ing. Bettar O. el Moctar, University of Duisburg-Essen, Germany
Author
Ebrahim Ghahramani
Chalmers, Mechanics and Maritime Sciences (M2), Marine Technology
Experimental and numerical study of cavitating flow around a surface mounted semi-circular cylinder
International Journal of Multiphase Flow,;Vol. 124(2020)
Journal article
A comparative study between numerical methods in simulation of cavitating bubbles
International Journal of Multiphase Flow,;Vol. 111(2019)p. 339-359
Journal article
Realizability improvements to a hybrid mixture-bubble model for simulation of cavitating flows
Computers and Fluids,;Vol. 174(2018)p. 135-143
Journal article
Analysis of the Finite Mass Transfer Models in the Numerical Simulation of Bubbly Flows
CAV2018 PROCEEDINGS,;(2018)
Paper in proceeding
Numerical simulation and analysis of multi-scale cavitating flows
Journal of Fluid Mechanics,;Vol. 922(2021)
Journal article
Cavitation is a phenomenon in which rapid changes of pressure in a liquid lead to the formation of small vapour-filled cavities in places where the pressure is relatively low. When subjected to higher pressure, these cavities collapse and can generate strong pressure waves. For decades cavitation has been the subject of numerous studies with the aim to reduce its undesirable consequences such as erosion, noise and efficiency loss in hydraulic machineries such as pumps and ship propellers. Recent advances in biomedical engineering, on the other hand, have caused significant interests in applying cavitating bubbles in cancer treatment, drug and DNA delivery and kidney stone lithotripsy, for instance. However, control of cavitation is still a challenge and a theoretical understanding is usually unachievable without significant simplifications. A main reason is that cavitating flows usually contain extensive ranges of length and time scales.
Thanks to recent improvements, today Computational Fluid Dynamics (CFD) is a reliable method to gain a more comprehensive understanding of the hydrodynamics of cavitation. The most common CFD models in literature can capture large cavities with sufficient accuracy. However, these models, which are called mixture models in this thesis, are limited in resolving small scale cavities. Apart from the mixture models, we have another group, which are called bubble models in this thesis. Bubbles models can represent the small scale cavities with sufficient accuracy, however, they are limited in accurate estimation of large vapour structures. In the current study, a new hybrid model is developed by coupling of a mixture and an improved bubble models, in order to capture an extensive range of cavity scales. The new model improves the prediction of cavity inception from micro-bubbles, its development to large cavities and the later collapse at small scales. Besides, the introduced improvements in bubble modelling, provides the possibility of more elaborate estimation of cavitation induced noise and erosion for future studies.
Thanks to recent improvements, today Computational Fluid Dynamics (CFD) is a reliable method to gain a more comprehensive understanding of the hydrodynamics of cavitation. The most common CFD models in literature can capture large cavities with sufficient accuracy. However, these models, which are called mixture models in this thesis, are limited in resolving small scale cavities. Apart from the mixture models, we have another group, which are called bubble models in this thesis. Bubbles models can represent the small scale cavities with sufficient accuracy, however, they are limited in accurate estimation of large vapour structures. In the current study, a new hybrid model is developed by coupling of a mixture and an improved bubble models, in order to capture an extensive range of cavity scales. The new model improves the prediction of cavity inception from micro-bubbles, its development to large cavities and the later collapse at small scales. Besides, the introduced improvements in bubble modelling, provides the possibility of more elaborate estimation of cavitation induced noise and erosion for future studies.
Development and experimental validation of computational models for cavitating flows, surface erosion damage and material loss (CaFE)
European Commission (EC) (EC/H2020/642536), 2015-01-01 -- 2019-01-01.
Subject Categories
Mechanical Engineering
Fluid Mechanics and Acoustics
Driving Forces
Sustainable development
Areas of Advance
Transport
Infrastructure
C3SE (Chalmers Centre for Computational Science and Engineering)
ISBN
978-91-7905-385-7
Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 4852
Publisher
Chalmers
Lecture hall HA1, Hörsalsvägen 4, Chalmers
Opponent: Prof. Dr.-Ing. Bettar O. el Moctar, University of Duisburg-Essen, Germany