讓結構設計同時達到安全性與經濟性一直以來都是工程師重要的議題，本研究探討基因演算法來設計平面鋼桁架橋的可行性，本研究與過去研究不同的地方在於：(1) 自行撰寫結構分析程式考量動態雙向移動活載與3種極限狀態、(2) 發展改良式基因演算法(添加初步搜尋、懲罰參數、菁英保留法和動態修正)，藉由簡單設計案例以窮舉設計結果來驗證改良式基因演算法可行性，然後由田口式實驗方式決定改良式基因演算法參數設定、最後(3) 從事接近實尺寸橋梁設計，比較改良式與一般式基因演算法之效率。

根據研究結果可得結論如下： (1)就簡單案例設計而言，使用窮舉法與使用改良式基因演算法所得結果相同，唯前者需超過30天來完成，而後者10次分析時間均約6分鐘、(2)就實尺寸橋梁設計而言，改良式基因演算法效率明顯高於一般式基因演算法，但是針對演算法所得仍然需要工程師進一步的專業判斷。

It has been always a challenge for engineers to find a balance between safety and economy in the structural design. To achieve this goal, the potential of using genetic algorithm (GA) to determine member sizes of 2D steel truss bridges is investigated in this study. Compared to the previous studies, this research is highlighted with (1) an self-developed structural analysis program that provides the author more flexibility to consider moving live load and the 3 limit states(2) the improved GA that include initial search, penalty parameter, elistic strategy and auto-tuning. The potential of using the improved GA is investigated though a simple example with the lightest but feasible member sizes obtained from the exhaustive method. After that, Taguchi method is conducted to determine the parameters in the improved GA, and (3) the efficiency of the improved GA is compared with the conventional GA through an approximately full-scale design case..

According to the research results, the following conclusions are drawn: (1) For a simple design case, the use of exhaustive method and the improved GA provide the same results. However, the analysis time spent by the exhaustive method is greater than 30 days while the average time spent by the improved GA for 10 analyses is about 6 minutes. (2) For the approximately full-scale design case, the efficiency of conventional GA is greatly improved by the improved GA but engineering judgement is still necessary to examine the results from the improved genetic algorithm.

AASHTO, 2012, AASHTO LRFD Bridge Design Specifications, 6th edition, American Association of State Highway and Transportation Officials, Washington, DC.

AISC, 2011, Steel Construction Manual, 14th edition, American Institute of Steel Construction, Chicago, IL.

Angelova, M., Roeva, O., amd Pencheva, T., 2015, InterCriteria Analysis of Crossover and Mutation Rates Relations in Simple Genetic Algorithm, Federated Conference on Computer Science and Information Systems (FedCSIS), pp. 419-424.

Cheng, J., 2010, Optimum design of steel truss arch bridges using a hybrid genetic algorithm, Constructional Steel Research, Vol. 66, pp. 1011-1017.

Dey, D. K., 2014, Mathematical Study OF Adaptive Genetic Algorithm (AGA) with Mutation and Crossover probabilities, Advanced Computer Technology, Vol. 3, NO. 5, pp. 765-768.

Holland, J. H., 1992, Adaptation in Natural and Artificial Systems, the MIT Press, Cambridge.

Lee, J. H., 2004, Local Buckling Behaviour and Design of Cold-formed Steel Compression Members at Elevated Temperatures, Ph. D Thesis, School of Civil Engineering, Queensland University of Technology, 347 pp.

LIN, W. Y., LEE, W. Y., and Hong, T. P., 2003, Adapting Crossover and Mutation Rates in Genetic Algorithms, Information Science and Engineering, Vol. 19, pp. 889-903.

Matlab(R2013a), 2013, Mathworks.

Minitab(v17), 2016, Minitab.

Risa(Demonstration v15), 2016, Risa.

SAP2000(v18), 2015, Computers & Structures,INC

S ̌es ̌ok, D., and Belevic ̌ius, R., 2008, Global optimization of trusses with a modified genetic algorithm, Civil Englneering and Management, Vol. 14, No. 3, pp. 147-154.

Taguchi, G., 1986, Introduction to Quality Engineering: Designing Quality into Products and Processes, Asian Productivity Organization.

Tog ̌an, V., and Dalog ̌lu, A. T, 2009, Bridge truss optimization under moving load using continuous and discrete design variables in optimization methods, Engineering & Materials Sciences, Vol. 16, pp. 245-258.

Weaver, Jr. W., Gere, J. M., 1990, Matrix Analysis of Framed Structures, 3th edition, Van Nostrand Reinhold, New York.

Yeh, I. C., 1999, Hybrid Genetic Algorithms for Optimization of Truss Structures, Computer-Aided Civil and Infrastructure Engineering, Vol. 14, pp. 199-206.

Yang, I. T., and Hsieh, Y. H., 2011, Reliability-based Design Optimization with Discrete Design Variables and Non-smooth Performance Functions: AB-PSO Algorithm, Automation in Construction, Vol. 20, pp. 610-619.

林采璇、王裕仁，2013，應用田口方法於基因演算法輸入參數設計-以工程品質查核標案選擇與委員指派組合為例，中工高雄會刊，第20卷，第3期，第44-53頁。

徐耀賜，2002，橋梁結構之基本功能，台北:全威圖書有限公司。

陸勇奇，2015，圖形處理器平行計算技術應用於影像處理及空間桁架結構最佳化之研究，博士論文，國立交通大學，新竹。

黃仲偉、陳北亭、吳啟誠，2010，適應性網格基因演算法於結構拓樸最佳化之應用，先進工程學刊，第五卷，第四期，第317-326頁。

張永康、周于文，2014，應用蜂群演算法於結構最佳化設計之研究，第九屆海峽兩岸航空太空學術研討會(頁351-359)，台北淡水。

楊子毅，2010，應用粒子群演算法調校層級分析法之權重矩陣以協助選商決策，碩士論文，國立台灣科技大學營建工程系，台北。

謝宜宏，2012，離散型可靠度最佳化設計之單目標與多目標演算架構，博士論文，國立台灣科技大學營建工程系，台北。