Inscrição na biblioteca: Guest
Portal Digital Begell Biblioteca digital da Begell eBooks Diários Referências e Anais Coleções de pesquisa
Computational Thermal Sciences: An International Journal
ESCI SJR: 0.249 SNIP: 0.434 CiteScore™: 0.7

ISSN Imprimir: 1940-2503
ISSN On-line: 1940-2554

Computational Thermal Sciences: An International Journal

DOI: 10.1615/ComputThermalScien.2015012264
pages 139-156

ASSESSMENT OF A LATTICE BOLTZMANN MODEL TO SIMULATE FLUID FLOWS WITH COMPLEX GEOMETRIES

Alissar Yehya
Civil Engineering and Geo-Environment Laboratory (LGCgE - EA 4515), Artois University, Faculty of Applied Sciences, Technoparc Futura, F-62400 Bethune, France; Lille University, Northern France, 59000 Lille, France
Hassane Naji
Civil Engineering and Geo-Environment Laboratory (LGCgE - EA 4515), Artois University, Faculty of Applied Sciences, Technoparc Futura, F-62400 Bethune, France; Lille University, Northern France, 59000 Lille, France
Laurent Zalewski
Civil Engineering and Geo-Environment Laboratory (LGCgE - EA 4515), Artois University, Faculty of Applied Sciences, Technoparc Futura, F-62400 Bethune, France; Lille University, Northern France, 59000 Lille, France

RESUMO

Lately, the lattice Boltzmann method (LBM) has proven to be a promising approach to solve complex fluid flows and transport phenomena. In the present contribution, the LB-BGK model, also called single relaxation time (SRT), is used to handle fluid flows in complex geometries. The model is first benchmarked via the 2D Poiseuille flow problem to conduct a discussion on its accuracy and performance. The accuracy of the LBM is weighted by several factors; among these we investigated the effect of the boundary conditions, the spatial resolution, the Mach number, and the choice of relaxation factors. It is found that the considered LBM is highly dependent on the physical problem, the numerical implementation, and the used parameters. However, a set of correlated optimal parameters is suggested to improve the accuracy and decrease both discretization and compressibility errors. In addition, we discussed the ability of the current model to handle complex geometries. The simulation of fluid past a circular or square cylinder in a confined geometry has been successfully achieved, even when curves are modeled as stairs. In light of the obtained results, we can state that the proposed model has high potential to handle problems with complex flows and geometries.