Artículo
Development of a parallelised fluid solver for problems with mesh interfaces and deforming domains
Aguerre, Horacio Javier
; Marquez Damian, Santiago
; Gimenez, Juan Marcelo
; Nigro, Norberto Marcelo
Fecha de publicación:
05/2018
Editorial:
Pergamon-Elsevier Science Ltd
Revista:
Computers & Fluids
ISSN:
0045-7930
Idioma:
Inglés
Tipo de recurso:
Artículo publicado
Clasificación temática:
Resumen
Numerical simulation of piston–cylinder problems such as internal combustion engines suffers from two difficulties related to discretization: the non-matching interfaces between cylinder and ports and the large deformation of the cylinder region. In this context, this paper presents a parallelized computational strategy, based on the Finite Volume method, where the mentioned issues are overcome. The non-matching discretization between regions is addressed with a new methodology which integrates the Arbitrary Coupled Mesh Interface (ACMI) method with the pseudo-supermesh approach to couple conservatively mesh interfaces with partial overlapping. For the large cylinder deformations, a layer addition/removal strategy is enhanced with mesh deformation to improve the handling of the mesh resolution. All methodologies are integrated and implemented for parallel computing obtaining a good scalability and computational efficiency. Specific examples are solved to analyse the performance of the new interface method and layering working separately and finally, the combination of both strategies is tested solving an internal combustion engine problem where the robustness of this dynamic mesh proposal is evaluated.
Archivos asociados
Licencia
Identificadores
Colecciones
Articulos(CIMEC)
Articulos de CENTRO DE INVESTIGACION DE METODOS COMPUTACIONALES
Articulos de CENTRO DE INVESTIGACION DE METODOS COMPUTACIONALES
Citación
Aguerre, Horacio Javier; Marquez Damian, Santiago; Gimenez, Juan Marcelo; Nigro, Norberto Marcelo; Development of a parallelised fluid solver for problems with mesh interfaces and deforming domains; Pergamon-Elsevier Science Ltd; Computers & Fluids; 168; 5-2018; 110-129
Compartir
Altmétricas