應用分散式類免疫演算法於 散熱片之最佳化設計

Optimization of Heat Sink Using Distributed Artificial Immune Algorithms

張慶彬(Ching-bin Zhang)

指導教授 ： 吳俊瑩

大同大學/工程學院/機械工程學系所/碩士(2004年)

摘要

近年來由於微處理機的效能與速度大大的提升，也因此導致許多散熱方面的問題。而在目前的電子工業中，散熱片被廣泛的運用於降低主要電子元件的溫度。而在散熱片的設計中，最困難的地方就是如何在最小的面積下將晶片上高密度的熱量帶走，而散熱片最佳化設計的目的就是設計出一個散熱效能很高的散熱片。

近年來許多學者對於生物系統演算法非常重視，像基因遺傳演算法、人工類免疫演算法、鳥群演算法與蟻群演算法。人體的類免疫系統是一廣泛而複雜的系統，它擁有多樣性、適應學習、記憶、分散式偵測等能力，而同時這些特質也是最佳化演算法的重要特性。因此我們利用類免疫演算法來做為我們散熱片最佳化的搜尋機制。

輕薄短小的電子商品是未來發展的趨勢，而產品所產生的高熱能以及設計上的諸多限制對於散熱設計人員來說是一項挑戰，因此工程輔助分析軟體（CAE）被廣泛的運用於分析複雜熱與流的物理現象，但是利用CAE軟體來分析是一項繁雜且耗時工作，因此運用費用低且效能高的叢集電腦來做分散式的平行分析是解決電子產品所產生複雜問題的最好辦法。

本研究中，我們利用分散式類免疫演算法整合CFDRC有限體積法之商業軟體去分析散熱片，以得到較好散熱效能的最佳形狀散熱片。為確認程式的正確性，我們先利用一些已知的測試函數做多值域最佳化的搜尋，以測試程式的架構及效能；確認程式的正確性後，利用此平行架構演算法求解散熱片的尺寸最佳化問題。經由研究的結果我們可以瞭解此架構能有效解決散熱鰭片的最佳化設計問題。

關鍵字

分散式類免疫演算法 ； 散熱片 ； 最佳化

並列摘要

The rapid advancement in technology of microprocessor design and manufacture has led to electronic thermal system designers to pay increasing attention to heat sink design and heat dissipation system layout. In the current electronics industry, the heat sinks are used extensively to provide cooling for key electronics components. The challenge in heat sink design becomes to remove heat from a small area with very high heat intensity for new microprocessor. The goal of optimal design of heat sink is to search the most adequate dimensions to achieve best performance of heat dissipation.
In last few years many researchers have pay attention to biologically inspired algorithms such as genetic algorithms, artificial immune algorithm, particle swarm and ant colony system. The immune system of human body is a comprehensive and complex system which has the capability of performing several tasks including adaptive learning, memory acquisition, generation of diversity, noise tolerance, and distributed detection. Those characteristics are also the important features of optimization algorithm and it is useful to transform the biological immune system into a searching algorithm for design optimization. The artificial immune algorithm is used as optimization engine for heat sink design in this study.
The trend of small and light requirements of electronics products challenge the capability of designer in cooling high heat intensity area with hard design constraints. The powerful computer-aided engineering(CAE) softwares have been used extensively in product design to achieve high quality specifications for complex designs with coupled fluid-thermal phenomenal. It is time-consuming job to analyze the complex fluid-thermal system using CAE software. The low cost and high performance cluster computers for parallel or distributed computation environment becomes a superior solution for designing complex electronics products.
The objective of optimization of heat sink in this study is to develop an integrated system by combining commercial finite volume software CFDRC with distributed artificial immune algorithm for extruded or cold forging fins. Some test functions are used to verify the accuracy, architecture and performance of the developed program in multimodal optimization problems. The developed program is then applied to the shape optimization of heat sink for light weight and high heat removal. From the results of this study it was shown that the integration of CFDRC software with distributed artificial immune algorithm for optimization of heat sink becomes affordable and effective.

並列關鍵字

Distributed Artificial Immune Algorithms ； Heat Sink ； Optimization

參考文獻

[1] Azar, K., McLeod, R. S., and Caron, R. E., ”Narrow Channel Heat Sink for Cooling of High Powered Electronic Components,” in Proc. 8th Annual IEEE Semi-Therm Symp., pp.12-19, 1992.

[4] Bar-Cohen, A., and Jelinek, M., ”Optimum Arrays of Longitudinal Fin in Convective Heat Transfer,” Heat Engineering, Vol. 6, pp.68-78, 1985.

[5] Bar-Cohen, A., ”Fin Thickness for an Optimal Natural Convection Array of Rectangular Fins,” ASME J. Heat Transfer, Vol. 101, pp.564-566, 1976.

[10] Iyengar, M., and Bar-Cohen, A., ”Least-Material Optimization of Vertical Pin-Fin, Plate-Fin and Triangular-Fin Heat Sinks in Natural Convective Heat Transfer,” IEEE International Society Conference on Thermal Phenomena, pp.295-302, 1998.

[11] Nowell, R. M., Jr., Bhavnani, S. H., and Jaeger, R. C., ”Effect of Channel Width on Pool Boiling from a Microconfigured Heat Sink,” IEEE Transactions on Components, Packaging, and Manufacturing Technology-Part A Vol. 18, No. 3, Sept. 1995.