Research Papers

An Artificial Neural Network Approach to Cooling Analysis of Electronic Components in Enclosures Filled With Nanofluids

[+] Author and Article Information
A. Kargar1

Engineering Faculty, Shahrekord University, P.O. Box 115, Shahrekord, Irankargar@ieee.org

B. Ghasemi

Engineering Faculty, Shahrekord University, P.O. Box 115, Shahrekord, Iranbehzadgh@yahoo.com

S. M. Aminossadati

School of Mechanical and Mining Engineering, The University of Queensland, Brisbane, Queensland 4072, Australiauqsamino@uq.edu.au


Corresponding author.

J. Electron. Packag 133(1), 011010 (Mar 10, 2011) (9 pages) doi:10.1115/1.4003215 History: Received January 05, 2010; Revised June 20, 2010; Published March 10, 2011; Online March 10, 2011

Computational fluid dynamics (CFD) and artificial neural network (ANN) are used to examine the cooling performance of two electronic components in an enclosure filled with a Cu-water nanofluid. The heat transfer within the enclosure is due to laminar natural convection between the heated electronic components mounted on the left and right vertical walls with a relatively lower temperature. The results of a CFD simulation are used to train and validate a series of ANN architectures, which are developed to quickly and accurately carry out this analysis. A comparison study between the results from the CFD simulation and the ANN analysis indicates that the ANN accurately predicts the cooling performance of electronic components within the given range of data.

Copyright © 2011 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

A schematic diagram of the physical model

Grahic Jump Location
Figure 2

Validation of the present work against Oztop and Abu-Nada (5)

Grahic Jump Location
Figure 3

Designed ANN architectures

Grahic Jump Location
Figure 4

Local Nusselt number along the left wall from CFD and ANN for pure fluid and nanofluid (Ra=106, H1=0.25, H2=0.75)

Grahic Jump Location
Figure 6

Streamlines (left) and isotherms (right) from CFD (—) and ANN (—) at different heat source locations (Ra=106, ϕ=0.05)

Grahic Jump Location
Figure 7

A comparison between CFD and ANN results in terms of the average Nusselt number versus the position of the top heat source (Ra=106, ϕ=0.05, H1=0.1)

Grahic Jump Location
Figure 5

(a) Variation in average Nusselt numbers of heat sources with solid volume fraction from CFD and ANN (Ra=106, H1=0.25, H2=0.75). (b) Variation in maximum temperatures of heat sources with solid volume fraction from CFD and ANN (Ra=106, H1=0.25, H2=0.75).




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In