臺灣博碩士論文加值系統

本研究目的為以基因演算法結合懲罰函數，建立適用二次多項式的非線性規劃之演算法。新演算法的運算步驟為(1)產生初始族群(2)計算適應值(內含懲罰函數)(3)選別(4)交配(5)突變(6)產生新族群(7)回到第(2)步(8)若符合終止條件則解碼並輸出。本試驗以MATLAB編輯程式，將新演算法寫成解二次多項式非線性規劃的軟體，演算法參數設為定值：世代數、母體大小、交配率、突變率分別設為500代或100代、100個、0.8 0.2，以醬油發酵條件的最佳化函數(此案例世代數為500代，其餘案例為100代)、單螺桿擠出加工水產飼料流速最大化、食品與生物工程及杏仁膠與阿拉伯膠之乳化能力等四案例研究做數值測試，探討此軟體之運算效率。新開發之程式應用於單螺桿擠出加工水產飼料流速最佳化、杏仁膠與阿拉伯膠之乳化力最佳化，皆有較原案例預測最適值更佳的目標函數值，尤以單螺桿擠出加工水產飼料流速最大化之改善13%最為顯著。新程式可替代AMPL等市售軟體運算二次非線性規劃，由於AMPL不公開程式碼，而MATLAB有公開程式碼，因此可進一步優化成對話框式的獨立程式，方便使用，且本程式不需更改起始點，運算效率較高。

Objective: In the study, we investigated genetic algorithm combining penalty function to develop an optimization program for RSM application of food and agriculture area.
Methods: The steps of the newly developed algorithm are (1) randomly producing the initial population, (2) calculation of fitness value (pseudo-objective function containing penalty method), (3) wheel selection, (4) crossover, (5) mutation, (6) new population generalization, (7) back to step (2), and (8) stop the circulation of algorithm and output the best solution if the termination criterion is met. The new program–genetic algorithm combing penalty method, which is to solve second degree polynomial model, was coded with MATLAB. The parameter used in the study: generations, population size, crossover rate, mutation rate were set by 100 or 500, 100, 0.8, 0.2, respectively. The new program is evaluated by several models of previous study about the relationship between the fermentation condition and quality of soy sauce, the optimization of biomass flow during single-screw extrusion cooking, the optimization of food and bioengineering processes and the assessment of emulsifying ability of almond gum and gum arabic.
Numerical test: the application of new program on the optimization of biomass flow and emulsification of almond gum and gum arabic are better than the previous studies predicted. The result of biomass flow have been raised 13% by the new program comparing to the result of previous study.
Conclusion: Newly developed program can be used as a substitute of AMPL to solve nonlinear programming problems of second degree polynomial model. The code of AMPL is undercover, while the one of MATLAB is opened. The new program can be further improved as an independent program for user friendly. In addition, repeatedly making initial guess is not necessary in the new program, thus the new program is more effective than traditional one.

摘要 II
目錄 V
圖目錄 VII
表目錄 VIII
壹、 前言 1
貳、 文獻回顧 3
1. 反應曲面法 3
2. 非線性規劃之最佳化 7
3. 基因演算法 8
參、 研究方法 12
1. 市售非線性規劃程式 12
2. MATLAB編碼 12
3. 基因演算法 (Genetic algorithm，GA) 13
4. 懲罰函數 (Penalty function) 18
5. 演算法流程 20
肆、 數值測試 23
1. 醬油初期發酵條件最佳化以提升產量 23
2. 單螺桿擠出加工水產飼料流速最大化 34
3. 最佳化食品及生物工程之製程 38
4. 以粒徑大小比較杏仁膠及阿拉伯膠之乳化能力 41
伍、 結論與未來工作 44
參考文獻 47

 王子慶，反應曲面法設計在尋求食品加工最適條件上之應用簡介。台灣省農業試驗所技術服務，1997。
 王怡舉、修乃華，非線性最優化理論與方法，2016。
 阮宣宏，評估二階段醬醪發酵法及最適化初期發酵條件以提高醬油品質。國立嘉義大學食品科學系碩士論文，2016。
 陳嘉珮，運用基因演算法及最佳化資源分配法求解隨機需求之長期車輛路線問題。國立交通大學運輸科技與管理學系碩士論文，2012。
 蔣安國、蔣安仁、丁敬哲、鐘健平，以模擬技術和統計方法建構專案排程之可靠度。管理科學研究 第一屆管理與決策2005年學術研討會特刊，2005。
 雙語詞彙，網址：
http://terms.naer.edu.tw/
 Bandyopadhyay Sukumar and Ranjan K. Rout, “Aquafeed extrudate flow rate and pellet characteristics from low-cost single-screw extruder,” Journal of Aquatic Food Product Technology, vol. 10, pp. 3–15, 2001.
 Bazaraa M. S., H. D. Sherali and C. M. Shetty, “Nonlinear Programming Theory and Algorithm,” 1993.
 Chen C.R., H.S. Ramaswamy, “Modeling and optimization of variable retort temperature (VRT) thermal processing using coupled neural networks and genetic algorithms,” Journal of Food Engineering, vol. 53, pp.209–220, 2002.
 Erdoğdu F., “Optimization in Food Engineering,” CRC Press, Inc. Boca Raton, FL, USA, 2009.
 Gen M. and R. Cheng, ” Genetic Algorithm & Engineering Design,” Wiley Publisher, 1997.
 Hicks C. R., Fundamental Concepts in the Design of Experiments, 1993.
 Holland, J. H.,” Adaptation in Natural and Artificial Systems,” 1975.
 Jinglie DOU, Sadiq T. and Shuryo N.,” Random-centroid optimization for food formulation,” Food Research International, vol. 26, pp. 27-37, 1993.
 Lee, H.; Yunus, R.; Juan, J.; Taufiq-Yap, Y.,” Process optimization design for jatropha-based biodiesel production using response surface methodology,” Fuel Processing Technology, vol. 92, pp. 2420–2428, 2011.
 Liu. X., X. Chen, W. Wu and G. Peng, “A neural network for predicting moisture content of grain drying process using genetic algorithm,” Food Control, vol.18, pp. 928-933, 2007.
 Myers R. H. and D. C. Montgomery, “Response surface methodology: process and product optimization using designed experiments,” Wiley Publisher, 1995.
 Nesrine Mahfoudhi, Mariarenata Sessa, Moncef Chouaibi, Giovanna Ferrari, Francesco Donsì, Salem Hamdi, “Assessment of emulsifying ability of almond gum in comparison with gum arabic using response surface methodology,” Food Hydrocolloids, vol. 37, pp. 49-59, 2014.
 Nguyen Xuan Hoang, Sophia Ferng, Ching-Hua Ting, Wei-Hua Huang, Robin Yih-Yuan Chiou, Cheng-Kuang Hsu,” Optimizing the initial moromi fermentation conditions to improve the quality of soy sauce,” Food Science and Technology, vol. 74, pp. 242-250, 2016.
 Rout R. K., “Studies on extrusion cooking in relation to shrimp feed processing,” Ph.D. Thesis, Indian Institute of Technology, Kharagpur, India, 1997.
 Ruckebusch C., “Chapter 12 Response surface methods,” Data Handling in Science and Technology, vol. 8, 1992.
 Shankar T. J., S. Sokhansanj, S. Bandyopadhyay and A. S. Bawa, “A case study on optimization of biomass flow during single-screw extrusion cooking using genetic algorithm (GA) and response surface method (RSM),” Food and Bioprocess Technology, vol. 3, pp.498–510, 2010.
 The MathWorks, Matlab 7 Getting Started Guide, 2009.
 Tumuluru J. S. and R. McCulloch, “Application of Hybrid Genetic Algorithm Routine in Optimizing Food and Bioengineering Processes,” Foods, vol. 5 (4), 76, 2016.
 Vadde K.K., V. R. Syrotiuk, and D. C. Montgomery, “Optimizing protocol interaction using response surface methodology,” IEEE Transactions on Mobile Computing, vol.5, pp. 627-639, 2006.
 Vimalashanmugam, K.; Viruthagiri, T.,” Response Surface Methodology Optimization of Process Parameters for Xylanase Production by Aspergillus fumigatus in SSF using Central Composite Design,” International Journal of Engineering Research and Applications, vol. 2, pp.277–287, 2012.