Abstract
A CFD simulation usually requires extensive computer storage and lengthy computational time. The application of artificial neural network models to thermal management of chips is still limited. In this study, the main objective is to find a neural network solution for obtaining suitable thickness levels and material for a chip subjected to a constant heat power. To achieve this aim a neural network is trained and tested using the results of the CFD program package Fluent. The back-propagation learning algorithm with three different variants, single layer and logistic sigmoid transfer function is employed in the network. By using the weights of the network, various formulations are designed for the output. The network has resulted in R 2 values of 0.999, and the mean% errors smaller than 0.8 and 0.7 for the training and test data, respectively. The analysis is extended for different thickness and input power values. Comparison of some randomly selected results obtained by the neural network model and the CFD program has yielded a maximum error of 1.8%, mean absolute percentage error of 0.55% and R 2 of 0.99994.
Similar content being viewed by others
Abbreviations
- ANN:
-
Artificial neural network
- Au:
-
Gold
- CFD:
-
Computational fluid dynamics
- Cu:
-
Copper
- FVM:
-
Finite volume method
- IC:
-
Integrated circuit
- LM:
-
Levenberg–Marquardt
- MAP:
-
Mean absolute percentage
- RMS:
-
Root mean squared
- SCG:
-
Scaled conjugate gradient
- Si:
-
Silicon
- C p :
-
Specific heat
- k :
-
Heat transfer coefficient
- q′′ :
-
Heat power
- R 2 :
-
Fraction of variance
- t 1 :
-
Thickness of lower plate
- t 2 :
-
Thickness of upper plate
- T :
-
Temperature (K)
- ρ :
-
Density
- x, y, z:
-
Rectangular coordinates
References
Bergles AE (2003) Evolution of cooling technology for electrical, electronic, and microelectronic equipment. IEEE Trans Compon Packag Technol 26(1):6–15
Metin Ertunc H (2006) Prediction of the pool boiling critical heat flux using artificial neural network. IEEE Trans Compon Packag Technol 29(4):770–777
Arik M, Bar-Cohen A (2003) Effusivity-based correlation of surface property effects in pool boiling CHF of dielectric liquids. Int J Heat Mass Transf 46:3755–3764
Fluent Inc. Product documentation
Toparli M, Sahin S, Ozkaya E, Sasaki S (2002) Residual thermal stress analysis in cylindrical steel bars using finite element method and artificial neural networks. Comput Struct 80:1763–1770
Tahir Ayata, Abdullah Cavuşoglu, Erol Arcaklıoğlu (2006) Predictions of temperature distributions on layered metal plates using artificial neural networks. Energy Convers Manag 47:2361–2370
Mihalakakou G (2002) On estimating soil temperature profiles. Energy Build 34:251–259
Çengel Y (1998) Heat transfer, a practical approach. McGraw-Hill, NY
Yan L, Haji-Sheikh A, Beck JV (1993) Thermal characteristics of two-layered bodies with embedded thin-film heat source. ASME J Electron Packag 115:276–283
Haji-Heikh A, Beck JV (2002) Temperature solution in multi-dimensional multi-layer bodies. Int J Heat Mass Transf 45:1865–1877
Haji-Heikh A, Beck JV, And Agonefer D (2003) Steady state heat conduction in multi-layer bodies. Int J Heat Mass Transf 46:2363–2379
Kakaç S, Yener Y (1993) Heat conduction, 3rd edn. Taylor and Francis, USA
Remsburg R (2001) Thermal design of electronic equipment. CRC Press, Boca Raton
Steinberg DS (1991) Cooling techniques for electronic equipment. Wiley, NY
Kraus AD (1995) Avram Bar-Cohen, design and analysis of heat sinks. Wiley, NY
Incropera FP (1999) Liquid cooling of electronic devices by single-phase convection. Wiley, NY
Dur FN (2007) Numerical analysis of conduction in multi-layered electronic equipment (in Turkish). Dissertation, Gazi University, Institute of Science and Technology, Dept. of Mechanical Engineering, Ankara, Turkey
Dizdar EN (2004) “Uygulamalı Olasılık ve İstatistik,” ABP Yayınevi&Matbaacılık (in Turkish)
Kalogirou SA (2000) Applications of artificial neural-networks for energy systems. Appl Energ 67:17–35
Bechtler H, Browne MW, Bansal PK, Kecman V (2001) New approach to dynamic modeling of vapourcompression liquid chillers: artificial neural networks. Appl Therm Eng 21:941–953
Mohandes M, Rehman S, Halawani TO (1998) Estimation of global solar radiation using artificial neural networks. Renew Energy 14(1–4):179–184
Matlab Version 6.0 Online Manual
Sözen A, Arcaklıoğlu E, Özkaymak M (2005) Turkey’s net energy consumption. Appl Energy 81:209–221
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ozsunar, A., Arcaklıoglu, E. & Nusret Dur, F. The prediction of maximum temperature for single chips’ cooling using artificial neural networks. Heat Mass Transfer 45, 443–450 (2009). https://doi.org/10.1007/s00231-008-0445-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00231-008-0445-x