鍋爐是現代工業製造過程中提供熱能、熱水、蒸汽等動力的主要來源，廣泛應用於汽電共生廠、化學工業製程加熱、造紙業、食品業之滅菌消毒製程、紡織業染整製程、電子製程清洗等等。其中，由於工業長時間的運轉需求與永續環保意識的提升，使得能源使用以及汙染排放成為評估鍋爐性能的重要指標。而影響鍋爐性能的重要因素除了鍋爐爐體、燃燒器以及熱交換器的設計以外，鍋爐運作仍會受到環境如溫度、氣體溫度等外部條件所影響，造成鍋爐性能的不穩定。因此，如何使鍋爐智慧化，提升鍋爐運作效率與降低污染排放量，是經濟與環保的重要議題。本論文以上述為出發點，於工業用熱水鍋爐建立數據驅動模型，針對鍋爐兩個重要指標：熱水加熱熱效率與氮氧化物排放量(NOx)進行模型建立，並使用基因演算法尋找優化參數組合，以期提升鍋爐性能。

Boiler is the major source to provide the thermal energy, hot water, steam, and power generation for the manufacturing processes of modern industries, which is widely used in steam-electricity cogeneration plant, process heating of chemical industry, paper manufacturing industry, sterilization processes of food industry, dyeing processes of textile industry, cleaning processes of electronic industry, etc. Hence, how to develop a smart boiler that could improve boiler efficiency and reduce emissions via the feedback control is an important economic and environmental issue. This paper has developed data-driven models from hot-water boiler, which has two important performance index of boiler, heat efficiency and NOx emission. Genetic Algorithm is applied to find the better control parameters of the boiler, which help improve the performance of the boiler.

〔1〕李政道、張義鴻及林杏秋，「鍋爐效率提昇」，絲織園地，第66期, 52-27頁，2010。
〔2〕H. Zhao, J. Shen, Y. Li, J. Bentsman, “Coal-fired utility boiler modelling for advanced economical low-NOx combustion controller design ”, Control Engineering Practice, Vol 58, pp. 127-141, January 2017.
〔3〕L. Xu, J. Yuan, “Online identification of the lower heating value of the coal entering the furnace based on the boiler-side whole process models ”, Fuel
Vol 161, pp. 68-77, December 2015.
〔4〕H. Rusinowski, W. Stanek, “Hybrid model of steam boiler ”, Energy, Vol 35, pp. 1107-1113, February 2010.
〔6〕C. Choeng Ryul, K., Chang, “Numerical investigation on the flow, combustion and NOx emission characteristics in a 500 MWe tangentially fired pulverized-coal boiler ”, Fuel, Vol 88, pp. 1720-1731, April 2009.
〔6〕R. Nikula, M. Ruusunen, K. Leiviskä, “Data-driven framework for boiler performance monitoring ”, Applied Energy, Vol 183, December 2016.
〔7〕J. Song, C. E. Romero, Z. Yao, B. He, “Improved artificial bee colony-based optimization of boiler combustion considering NOX emissions, heat rate and fly ash recycling for on-line applications ”, Fuel, Vol 172, pp. 20-28, May 2016.
〔8〕G. Li, P. Niu, H. Wang, Y. Liu, “Least Square Fast Learning Network for modeling the combustion efficiency of a 300WM coal-fired boiler ”, Neural Network, Vol 51, pp. 57-66, March 2014.
〔9〕Q. Li, G. Yao, “Improved coal combustion optimization model based on load balance and coal qualities ”, Energy, Vol 132, pp. 204-212, August 2017.
〔10〕Z. Tang, Z. Zhang, “The multi-objective optimization of combustion system operations based on deep data-driven models”, Energy, Vol 182, pp. 37-47, September 2019.
〔11〕周志儒、李建利，「用統計方法於現場高壓蒸氣鍋爐提昇能源效率且降低氮氧化物排放之製程最適化研究」，行政院國家科學委員會專題研究計畫，2004。
〔12〕黃逸芳，「氫燃燒器與低氮氧化物燃燒器實作研究」，國立中央大學，碩士論文，民國95年。
〔13〕B. G. Miller, Fossil fuel emission control technologies : Stationary heat and power systems., Butterworth–Heinemann, United Kingdom, 2015.
〔14〕A. Fischer, C. Igel, “An Introduction to Restricted Boltzmann Machines ”, Vol 7441, pp. 14-36, Springer CIARP, 2012.
〔15〕G. E. Hinton. “Deep belief networks ”, Scholarpedia, Vol 4, pp. 5947, 2009.