Parallel numerical computation of mixed convection in a square enclosure with multiple heat sources on Anucluster Mixed convection in a square cavity

The coupled nonlinear partial differential equations modelling the influence of multiple heat sources on the mixed convection process in a fluid saturated vertical porous square enclosure have been solved numerically by Galerkin finite element method in conjuction with the Krylov Solver BICGM. Detai...

Full description

Saved in:
Bibliographic Details
Published in:International journal of advances in engineering sciences and applied mathematics Vol. 7; no. 3; pp. 96 - 105
Main Authors: Kumar, B. V. Rathish, Nigam, Mohit, Kumari, Priti, Sharma, Parul
Format: Journal Article
Language:English
Published: New Delhi Springer India 01.09.2015
Subjects:
Engineering
Mathematical and Computational Engineering
Finite element method
Mixed convection
Darcy
Porous medium
ISSN:0975-0770, 0975-5616
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The coupled nonlinear partial differential equations modelling the influence of multiple heat sources on the mixed convection process in a fluid saturated vertical porous square enclosure have been solved numerically by Galerkin finite element method in conjuction with the Krylov Solver BICGM. Detailed parallel numerical computations, based on the message passing libraries from ANULIB, have been carried out on an eight noded ANUCLUSTER to assess the influence of flow governing paremeters such as Rayleigh number and suction/injection velocity on mixed convection process. The obtained results have been analysed for heat and mass transfer in the form of isotherm, streamline and velocity vector plots. Local, cumulative and global heat fluxes have been calculated in the form of corresponding versions of Nusselt number. Computations have also been carried out for the case natural convection due the multiple hot sources. Nusselt numbers from mixed convection process are compared with those from natural convection process. Efficiency and speedup of the parallel computations on ANUCLUSTER have also been assessed and presented.
ISSN:0975-0770
0975-5616
DOI:10.1007/s12572-015-0131-4