臺灣博碩士論文加值系統

近年來人們對於靈感來自生物所衍生出來的仿生演算法，越來越感興趣；一些仿生演算法像是類神經網路、基因遺傳演算法、類免疫演算法以及蟻群演算法等，皆有許多相關的文章研究探討。仿生演算法是一種廣泛且複雜的系統，其中尤以類免疫演算法具備了適應性學習、記憶性、多樣性、誤差容忍以及分散式搜尋等特性；而這些特性正是類免疫演算法能有效地轉為最佳化設計搜尋演算法的特徵。
由於電腦的迅速發展，電腦輔助工程分析（Computer-Aided Engineering, CAE）的軟體，變得功能更加強大；且在時間競爭的壓力下，成本與品質間的考量，對產品製造的影響已逐漸加劇；運用CAE分析軟體，在短時間內設計出低成本與高品質的產品已變得十分重要。但若能把CAE分析軟體整合至一強健且有效率的搜尋引擎，這將有助於提升複雜且有高精度要求的設計。整合CAE分析軟體於最佳化設計可節省許多時間；一部個人電腦即可完成一個也許要超過一個月才能完成的最佳化設計工作。但為了要更進一步縮短設計時程與降低成本，平行及分散式架構的叢集電腦將是唯一的選擇。
本研究在視窗作業系統下運用C++程式語言及WinkSock API，於分散式架構的叢集電腦環境上，開發多值域類免疫演算法程式。同時也整合商用分析軟體ANSYS應用於工程最佳化。本研究一開始採用數個測試方程式測試與驗證程式之正確性與執行效率。接著將本程式應用於多值域二次元結構拓樸最佳化設計，最後探討程式執行之結果與效能。證明分散式多值域類免疫演算法整合商用分析軟體，能於短時間內，協助研發複雜且低成本高品質的設計，進而提升產品的競爭力。

In last few years there is a great increase of interest in learning biologically inspired systems. Some biologically inspired algorithms such as artificial neural network, genetic algorithms, artificial immune algorithm and ant colony system are emphasized in many published papers. The biologically inspired system is a comprehensive and complex system. The artificial immune algorithm specially has capability of performing several tasks including adaptive learning, memory acquisition, generation of diversity, noise tolerance, and distributed detection. Those characteristics are also the system feature of optimization algorithms and it is useful to transform the biological system into searching algorithm for design optimization.
Due to the development of computer, the computer-aided engineering(CAE) software becomes powerful and friendly. The pressure of competition among time, cost and quality is increased for product. It becomes important to design product using CAE software for low cost and high quality product in a short period of time. The integration of CAE software with a robust and efficient search engine becomes important for improving complex design in quality and precision. It is time-consuming work for using CAE software for optimization search and may take more than one month to finish a single design optimization job by using just single personal computer. In order to reduce the design period and cost, the only way is to use cluster PCs for parallel or distributed computation environment.
A distributed artificial immune algorithm will be developed in this study for Windows operation systems using TCP/IP, winksock and C++ language. The ANSYS software will be integrated with distributed artificial immune algorithm for engineering optimization. Some test functions are used first to verify the correctness and performance of developed program. Then multi-modal topological optimization of structure will be used to prove the performance of distributed artificial immune algorithm. This also shows that the integration of CAE software with distributed artificial immune algorithm can really help the industry to develop low cost, and high quality complex design in short design period. The design of satisfied product will become faster and easier. The design compatibility, product quality, and cost control will be improved for competition.

中文摘要…….………......…………………………………………..……i
英文摘要…….…………….........………………………….…………….ii
誌謝……..………..……………………………...………………………iii
目錄….…………………......………………………………...………….iv
圖目錄………………….......…………………………………………. viii
表目錄……………….......………………………………..……………..ix
第一章 緒論…………………………........…………………..………….1
1.1 前言………..…………………………………………………1
1.2 文獻回顧….....…………………………………………...…..3
1.2.1 拓樸最佳化設計…......…....…....…..…....…......…....…....3
1.2.2 分散式計算…....…....…....…..……..…....…......…....…....5
1.2.3 類免疫演算法…....…....….....…....…..…..….…....…....…6
1.3 研究動機與目的....……………………………………….….7
第二章 多值域類免疫演算法….………………………….…………….9
2.1 簡介…………………..……………………….....…………..9
2.2 生物免疫反應之介紹……….………………….…...……..10
2.3 類免疫演算法之特點……….…………..………..………..16
2.4 類免疫演算法之組織架構………………………………...17
2.5 測試方程式之驗證………………………………………...23
第三章 分散式多值域類免疫演算法應用於結構拓樸最佳化設計之架構…………………………………………………………..…..32
3.1 簡介…………………………..………….…………………32
3.2 有限元素分析之介紹………………………….…………..33
3.2.1 有限元素法之簡介……………………………………..33
3.2.2 有限元素分析之程序…………………………………..34
3.2.3 有限元素法理論分析…………………………………..34
3.3 多值域類免疫演算法應用於結構拓樸最佳化設計..…….39
3.3.1 結構編碼………………………………………………..39
3.3.2 連續性修正……………………………………………..40
3.3.3 輕重鍵定義……………………………………………..42
3.3.4 親和度之計算設定……………………………………..44
3.4 整合商用軟體ANSYS之架構介紹………………………45
3.5 分散式多值域類免疫演算法之架構介紹………………...47
3.5.1 分散式計算之簡介……………………………………..47
3.5.2 分散式多值域類免疫演算法之架構…………………..48
第四章 多值域結構拓樸最佳化設計實例………..........…………….51
4.1 類免疫演算法應用於結構拓樸最佳化設計之執行結果……………….. …………….. ……………... .…………51
4.2 分散式計算架構之效能探討……………………………...57
4.3 整合ANSYS與分散式多值域類免疫演算法於結構拓樸最佳化設計…………………………………...……………...60
第五章 結論與未來展望..............................................................…….64
5.1 結論………………………………………………………...64
5.2 未來展望…………………………………………………...65
參考文獻…............................…………………….…………………….66
附錄 程式介面..................................................................................69
作者簡介……………………….........................……………………….70

[1] M. Cerrolaza, W. Annicchiarico and M. Martinze, Optimization of 2D Boundary element models using B-splines and genetic algorithms, Engineering Analysis with Boundary, Elements, Volume 24, pp.427-440, 2000.
[2] M. P. BendsÆe and N. Kikuchi, Generating optimal topologies in structural design using a homogenization method, Computer Methods in Applied Mechanical and Engineering, Volume 71, pp.197-224, 1988.
[3] H. P. Mlejnek, Some aspects of the genesis of structures, Structural Optimization, Volume 5, pp.64-69, 1992.
[4] C. D. Chapman, K. Saitou and M. J. Jakiela, Genetic algorithms as an approach to configuration and topology design, Journal of Mechanical Design, Volume 116, pp.1005-1011, 1994.
[5] M. J. Jakiela, Colin Chapman and James Dude, Continuum structural topology design with genetic algorithms, Comput. Methods Appl. Mech. Engrg, Volume 186, pp.339-356, 2000.
[6] J. Rodriguez and J. R. Veiazquez, Optimizing the shapes of structures via a rule-based computer program, Computers in Mechanical Engineering, pp.20-27, July 1985.
[7] T. Sterling, D. Becker, D. Savarese, BEOWULF: A parallel workstation for scientific computation, Proceedings of the 1995 International Conference on Parallel Processing (ICPP), August 1995.
[8] T. Sterling, D. Savarese , D. Becker, Communication overhead for space science applications on the Beowulf Parallel Workstation, Proceedings of the 4th IEEE Symposium on High Performance Distributed Computing (HPDC), August 1995.
[9] C. Reschke, T. Sterling and D. Ridge, A design study of alternative network topologies for the Beowulf Parallel Workstation, Proceedings of the 5th IEEE Symposium on High Performance and Distributed Computing, August 1996.
[10] D. Ridge, D. Becker and P. Merkey, Beowulf: Harnessing the power of parallelism in a Pile-of-PCs, Proceedings of IEEE Aerospace, 1997.
[11] M. S. Warren, John K. Salmon, Pentium ProInside: (I). A Treecode at 430 Gigaflops on ASCI Red (II). Price/Performance of $50/Mflops on Loki and Hyglac, Los Alamos National Labortory and Center for Advanced Computing Research, Caltech, 1997.
[12] M. Snir, S. Otto, MPI: The Complete Reference, The MIT Press, Cambridge, Massachusetts, 1996.
[13] Ishida, Y., Hirayama, H., Fujita, H., Ishiguro, A. and Mori, K. “Immunity-Based Systems—Intelligent Systems by Artificial Immune Systems”, Corona Pub. Co. Japan(in Japanese),1998.
[14] Dasgupta, D. “Artificial Immune Systems and Their Applications”, Springer-Verlag, 1999.
[15] de Castro, L. N., and von Zuben, F. J., ”Artificial Immune Systems：Part I-Basic Theory and Applications,” Technical Report TR-DCA 01/99, 1999.
[16] John E. Hunt and Denise E. Cooke, Learning using an artificial immune system, Journal of Network and Computer Applications, pp. 189-212, 1996.
[17] Fukuda, T., Mori, K. & Tsukiama, M., Parallel search for multi-modal function optimization with diversity and learning of immune algorithm, Artificial Immune Systems and Their Application, Springer-Verlag, pp. 210-220,1999..
[18] Han Shenglian, Ni Meng, and Ge Wancheng, Problems in GA and necessarity of importing immune function, Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress on, Volume 1, pp.542-544, 2000.
[19] Jiao Licheng and Wang Lei, A novel genetic algorithm based on immunity, Systems, Man and Cybernetics, Part A, IEEE Transactions on, Volume 30 Issue 5, pp.552-561, Sept. 2000.
[20] Lei Wang and Licheng Jiao, A novel genetic algorithm based on immunity, Circuits and Systems, 2000. Proceedings. ISCAS 2000 Geneva. The 2000 IEEE International Symposium on, Volume 5, pp.385-388, 2000.
[21] Gasper, A. and Collard, P., From GAs to artificial immune systems: improving adaptation in time dependent optimization, Evolutionary Computation, CEC 99. Proceedings of the 1999 Congress on, Volume 3, pp.1859-1866, 1999.
[22] 陳如斌, 二次元之結構外型最佳化： 類免疫演算法之應用 碩士論文, 大同大學機械工程研究所, 2003.
[23] Yang, R. J., and Chou, J. N., Optimal topology design using linear programming, Computers and Structures, Volume 52, pp.265-275, 1994.