鋼鐵是現今社會中不可或缺的一部份，整個煉鋼過程需將原物料處理成鐵水，再加工成鋼，並將鋼凝固成板坯。通常煉鋼分為三個步驟：煉鋼、精煉、連續鑄造，而在精煉與連續鑄造階段，如果半成品工件無法及時處理，則需要保持溫度，會增加加熱成本，因此有效的煉鋼排程對於提高鋼鐵廠的生產效率非常重要。因為煉鋼時的各種已知與未知變化因素眾多，例如：加工時間、移動時間、閒置時間、機台損壞、人為破壞、天災等，以人工排程頗不實際，因此本研究應用兩種演算法，包括：灰狼演算法(Grey Wolf Optimizer, GWO)與粒子群演算法(Particle Swarm Optimization, PSO)，來規劃煉鋼之排程。本研究以某鋼鐵廠為例，提出一個新的編碼方式並使用兩種人工智慧演算法進行求解，結果顯示PSO求解品質優於GWO，而GWO求解速度優於PSO。

Steel is one of the most important products in the world today. The process of processing molten iron into steel and solidifying the steel into slabs is a challenge for steel mills. There are three main steps of steelmaking: steelmaking, refining and continuous casting. If the workpiece cannot be processed in time in the refining and continuous casting stage, the temperature needs to be maintained which will increase the heating cost. Therefore, effective steelmaking scheduling is important to improve the production efficiency of the steel plant. Considering the various changes in steelmaking, manual scheduling cannot afford so many changes in time. So this study uses Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO) to solve for the steelmaking schedule. This study takes a steel plant in China as an example to analyze and compare the benefits of these two algorithms. Comparing the results, although GWO is faster than PSO, PSO outperforms GWO.

摘要......i
Abstract......ii
誌謝......iii
目錄......iv
表目錄......vi
圖目錄......vii
第一章 緒論......1
1.1 研究背景與動機......1
1.2 研究目的......1
1.3 研究方法與步驟......1
1.4 論文架構......2
第二章 文獻探討......4
2.1 煉鋼相關文獻......4
2.1.1 煉鋼......5
2.1.2 精煉......5
2.1.3 連續鑄造......6
2.2 元啟發式算法......7
2.3 灰狼演算法......7
2.3.1 灰狼演算法介紹......7
2.3.2 灰狼演算法公式......8
2.3.3 GWO步驟......10
2.4 粒子群演算法......10
2.4.1 粒子群演算法介紹......10
2.4.2 粒子群演算法步驟......12
2.5 田口法(參數設定)......13
2.5.1田口法介紹......13
2.5.2 品質損失函數(Loss Function)......13
2.5.3 直交表......14
2.5.4 信號雜音比(S/N比)......15
2.5.6 田口法步驟......16
第三章 研究問題與研究方法......17
3.1 煉鋼排程問題......17
3.2 編碼方法......18
3.3 演算法參數設定......21
3.3.1 GWO之參數設定......22
3.3.2 PSO之參數設定......23
第四章 測試問題與結果分析......25
4.1 參數設定及設備環境......25
4.2 測試問題......25
4.3 測試結果......26
4.4 統計檢定......36
第五章 結論......41
5.1 結論......41
5.2 未來研究方向......41
參考文獻......42
附錄A......44
附錄B......70
Extended Abstract......78

[1] 謝之駿(1996)，百鍊成鋼－煉鐵煉鋼的真功夫，中興大學材料科學與工程研究所，台灣。
[2] 吳展維(2017)，應用翠思手法改善盛鋼桶流鋼嘴座磚之壽命－以某鋼鐵廠為例，國立高雄應用科技大學工業工程與管理系碩士在職專班，碩士論文。
[3] 黃清龍(2014)，二次精煉最佳脫硫率之研究，南開科技大學電子工程研究所，碩士論文。
[4] 吳舒昌(2015)，高猛鋼連續鑄造之高溫熱機械性質量測與CALPHAD熱力學解析，國立成功大學材料科學及工程學系，碩士論文。
[5] Bogdan Filipic (2015), “In search for an efficient parameter tuning method for steel casting,” Retrieved from https://www.researchgate.net
[6] C. Dhaenens (2016), “Optimization and Big Data,” Metaheuristics for Big Data, pp. 8-10.
[7] M. Niu (2016), “A novel hybrid decomposition-and-ensemble model based on CEEMD and GWO for short-term PM2.5 concentration forecasting”, Atmospheric Environment, vol. 134, pp. 168-180.
[8] Mirjalili, S. & Saremi, S. (2016), “Multi-objective grey wolf optimizer: A novel algorithm for multi-criterion optimization”, Expert Systems with Applications, vol. 47, pp, 106-119.
[9] Mirjalili, S. & Saremi, S. & Lewis, A. (2014), “Gery Wolf Optimizaer”, Advances in Engineering Sorfware, vol. 69, pp. 46-61.
[10] Saxena, P. & Kothari, A. (2016), “Optimal Pattern Synthesis of Linear Antenna Array Using Gery Wolf Optimization Algorithm”, vol. 2016, doi: 10.1155/2016/1205970.
[11] Nahak, N. & Malliick, R. K. (2017), “Damping of power system oscillations by a novel DE-GWO optimized dual UPFC controller”, Engineering Science and Technology, an International Journal, vol. 20, pp. 1275-1284.
[12] 李伯鴻(2007)，台灣鋼鐵業之競爭策略-以中鋼為例，國立中山大學企業管理碩士研究所，碩士論文。
[13] 徐誠佑 (2006)，粒子群演算法於離散最佳化問題之研究，國立陽明交通大學工業工程與管理學系，碩士論文。
[14] 陳士杰 (2017)，粒子群演算法基礎，國立聯合大學資訊管理學系，機器學習課程。
[15] Xing, Z. & Zhu, J. & Qin, Y. & Jia, L. (2022), “Energy consumption optimization of tramway operation based on improved PSO algorithm”, Energy, vol. 258, pp. 124848.
[16] 謝采蓉 (2022)，應用三種人工智慧演算法於偏光板原物料切割排程作業，國立虎尾科技大學工業管理系工業工程與管理碩士班，碩士論文。
[17] 童超塵 (2020)，田口品質工程，國立雲林科技大學工業工程研究所，品質與可靠度工程實驗室。
[18] Liao, C. N. (2010), “Supplier selection project using an integrated Delphi, AHP and Taguchi loss function”, ProbStat Forum, vol. 03, pp. 118-134.
[19] Bhamare, S. & Yadav O. P. & Rathore A. P. (2009), “A Hybrid Quality Loss Function-Based Multi-Objective Design Optimization Approach”, Quality Engineering, vol. 21, pp.227-289.
[20] Abdu, H. M. & Gadallah, M. (2015), “Modeling and Optimization of Laser Cutting Operations”, Manufacturing Review, vol. 2, pp.1-15, doi: 10.1051/mfreview/2015020.
[21] 黃禹璇 (2021)，應用人工智慧演算法於多型態遊樂園途程問題：以義大遊樂世界為例，國立虎尾科技大學工業管理系工業工程與管理碩士班，碩士論文。
[22] 簡榜萱 (2017)，田口直交表實驗法進行稜鏡光學系統之最佳化設計，高苑科技大學，碩士論文。
[23] Luo, Q. & Rao, Y. & Peng, D. (2022), “GA and GWO algorithm for the special bin packing problem encountered in field of aircraft arrangement”, Applied Soft Computing, vol. 114, pp. 108060.
[24] Long, W. & Jiao, J. & Liang, X. & Tang, M. (2018), “An exploration-enhanced grey wolf optimizer to solve high-dimensional numerical optimization”, Engineering Applications of Artificial Intelligence, vol. 68, pp. 63-80.
[25] Ghalambaz, M. & Yengejeh, R. J. & Davami, A. H. (2021), “Building energy optimization using Grey Wolf Optimizaer (GWO)”, Case Studies in Thermal Engineering, vol. 27, pp. 101250.
[26] Fan, X. & Gao, Z. & Ji, W. & Lin, W. & Wang, J. (2021), “Thermodynamic optimization with multi objectives and parameters for liquid air energy storage system based on the particle swarm optimization (PSO)”, Journal of Energy Storage, vol. 41, pp. 102878.
[27] Nik, A. & Nejad, F. M. & Zakeri, H. (2016), “Hybrid PSO and GA approach for optimizing surveyed asphalt pavement inspection units in massive network”, Automation in Construction, vol. 71, pp. 325-345.
[28] Xu, L. & Cao, M. & Song, B. (2022), “A new approach to smooth path planning of mobile robot based on quartic Bezier transition curve and improved PSO algorithm”, Neurocomputing, vol. 473, pp. 98-106.