Numerical approach to study bubbles and drops evolving through complex geometries by using a level set – Moving mesh – Immersed boundary method

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Vol. 349; pp. 662 - 682
• Main Authors: Gutiérrez, E., Favre, F., Balcázar, N., Amani, A., Rigola, J.
• Format: Journal Article Publication
• Language: English
• Published: Elsevier B.V 01.10.2018
• Subjects: Anàlisi numèrica; Arbitrary Lagrangian-Eulerian formulation; Complex geometries; Immersed boundary method; Level set method; Matemàtiques i estadística; Multiphase flow; Numerical analysis; Unstructured meshes; Àrees temàtiques de la UPC; Arbitrary Lagrangian-Eulerian formulation; Unstructured meshes; Immersed boundary method; Multiphase flow; Complex geometries; Level set method
• ISSN: 1385-8947; 1873-3212
• DOI: 10.1016/j.cej.2018.05.110

[Display omitted] •A new formulation is proposed to study drops/bubbles in complex geometries.•The multiphase domain is successfully tackled using a conservative level set method.•The simulation domain is optimized by using a moving mesh.•Inner and intricate boundaries are handled by using an immersed boundary method.•Extensive numerical tests were conducted in order to validate the proposed method. The present work proposes a method to study problems of drops and bubbles evolving in complex geometries. First, a conservative level set (CLS) method is enforced to handle the multiphase domain while keeping the mass conservation under control. An Arbitrary Lagrangian-Eulerian (ALE) formulation is proposed to optimize the simulation domain. Thus, a moving mesh (MM) will follow the motion of the bubble, allowing the reduction of the computational domain size and the improvement of the mesh quality. This has a direct impact on the computational resources consumption which is notably reduced. Finally, the use of an Immersed Boundary (IB) method allows to deal with intricate geometries and to reproduce internal boundaries within an ALE framework. The resulting method is capable of dealing with full unstructured meshes. Different problems have been studied to assert the proposed formulation, both involving constricting and non-constricting geometries. In particular, the following problems have been addressed: a 2D gravity-driven bubble interacting with a highly-inclined plane, a 2D gravity-driven Taylor bubble turning into a curved channel, the 3D passage of a drop through a periodically constricted channel, and the impingement of a 3D drop on a flat plate. Good agreement was found for all these cases study, which proves the suitability of the proposed CLS + MM + IB method to study this type of problems.