Evaluation of rigid body force in liquid sloshing problems of a partially filled tank: Traditional CFD/SPH/ALE comparative study

• Published in: Ocean engineering Vol. 236; p. 109556
• Main Authors: Cai, Zhemin, Topa, Ameen, Djukic, Luke P., Herath, Manudha T., Pearce, Garth M.K.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15-09-2021
• Subjects: Ansys fluent; Arbitrary Lagrangian–Eulerian; Liquid sloshing; Ls-Dyna; Smooth particle hydrodynamics; Turbulence; Volume of fluid; Turbulence; Ansys fluent; Arbitrary Lagrangian–Eulerian; Smooth particle hydrodynamics; Liquid sloshing; Volume of fluid; Ls-Dyna
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2021.109556

The rigid body force (RBF) generated by fluid movement inside a partially filled tank is of great importance for many tank structures transporting liquid under dynamic conditions. However, accurate prediction of such forces presents a challenge for the designer due to the high level of nonlinearities involved. In this paper, three fundamentally different approaches; Eulerian Volume of Fluid (VOF) approach with three common turbulence models; Smoothed Particle Hydrodynamics (SPH); and Arbitrary Lagrangian–Eulerian (ALE) are used to simulate liquid sloshing responses. The three approaches are implemented across two commercial simulation packages (ANSYS Fluent and Ls-Dyna) and evaluated using the experimental measurement from Yan (2008). A detailed quantitative comparative study on the RBF among various numerical methods are carried out, while a good agreement is found among the converged solution of using the LES turbulence model in Ansys Fluent; SPH and ALE in Ls-Dyna; while the ICFD model presents slower convergence to the experiment result compared to the other solvers. On other hand, all the numerical approaches fail to accurately predict the maximum absolute RBF while the wave is sharply separated from the main domain and this force overprediction/underprediction is dependent on element size, time step size and fluid model. Impulse, rather than absolute maximum RBF is proposed as the main computational characteristic to validate the performance of the baffled/unbaffled transport tanks under fluid sloshing.
•The RBF generated by fluid movement has been simulated in Fluent, ICFD ALE and SPH method in Ls-Dyna.•Overpredict of RBF occurs when fluid separates from main domain with high speed.•This overpredict is element size, time step size, and fluid model dependent.•Impulse rather than maximum RBF is recommended to validate fluid performance.