摘要

本研究主要了解ANAMMOX 反應於EABD-AO之效益和ANFIS之預測目的於增進污水處理廠處理之效益。主要分為兩大部分：1. ANFIS: (1) 應用相關系數方法篩選系統參數探討應用適應性模糊推論系統 ( adaptive network based fuzzy inference system, ANFIS)並建立模擬模型。(2) 探討ANFIS對工業區廢水廠出流水水質之預測效能。2. ANAMMOX: (1) 建立新穎之去除碳、氮、磷併同ANAMMOX 程序。(2)利用EABD-AO 活性污泥系統進而了解其活性污泥微生物功能族群之反應常數。研究結果顯示， ANFIS 對於工業區廢水廠出流水水質之預測結果： ANFIS之 SSeff、CODeff、pHeff預測結果最佳之平均百分誤差值 ( mean absolute percentage error, MAPE) 分別為13.88%、9.47%、2.11%；相關係數 ( correlation coefficient, R )分別為0.41、0.36、0.80；經研究結果所示ANFIS之MAPE與 R 值皆屬於高度相關性，顯示模擬值與實際值將近一致。而模廠活性污泥微生物功能族群之反應常數驗證EABD-AO模廠於添加無機碳後ANAMMOX反應有助於模廠之去除率，於批次筒試驗及ANAMMOX計量計算中更加證實其存在性，有助於了解其反應所消耗的能量常數。

Abstract

This study aims to investigate the effectiveness of applying ANAMMOX in EABD-AO and the ANFIS prediction purposes for enhancing the effectiveness of sewage treatment plants. The paper is divided into two parts: 1) ANFIS: (1) to apply the correlation coefficient method to screen system parameters for discussion of applying ANFIS (adaptive network based fuzzy inference system) and establish the simulation model; (2) to discuss the ANFIS’ efficiency in predicting the quality of industrial wastewater plant effluents; 2) ANAMMOX: (1) to establish the innovative removal procedures of carbon, nitrogen and phosphorus and the ANAMMOX program; (2) to use the EABD-AO active sludge system to understand the response constant of the active sludge microbial functional groups. The results showed that, the prediction results of ANFIS regarding the industrial wastewater plant effluents are: the optimal MAPE (mean absolute percentage error) of SSeff, CODeff, pHeff as predicted by ANFIS are 13.88%, 9.47%, 2.11%, respectively; CR (correlation coefficient) are 0.41, 0.36, 0.80 respectively. The results suggested that, the MAPE and R value of ANFIS are highly correlated, indicating that the simulated values and the actual values are almost consistent. The response constant of the simulated active sludge microbial functional groups has confirmed that the ANAMMOX reaction after the addition of inorganic carbon in the EABD-AO simulated plant can help the removal rate of the plant. The batch barrel test and ANAMMOX calculation also confirm its existence, thus help to understand the constant of the energy consumed in reaction.

目錄
摘要 I
誌謝 V
目錄 VII
表目錄 .X
圖目錄 ..XII
第一章 前言 1
1.1 研究緣起 1
1.2 研究目的 4
1.2.1 ANFIS 4
1.2.2 ANAMMOX 6
第二章 文獻回顧 7
2.1 ANFIS 7
2.1.1 模糊理論 7
2.1.2 類神經網路之發展 7
2.1.3 類神經網路的優缺點 10
2.1.4 適應性模糊類神經推論系統之發展 12
2.1.5 類神經網路和模糊系統的共通點及差異性 18
2.2 ANAMMOX 23
2.2.1國外缺氧/厭氧/好氧處理程序的發展現況 23
2.2.2 國外ANAMMOX 程序的發展現況 23
2.2.3 國內缺氧/厭氧/好氧處理程序及ANAMMOX 程序的發展現況 25
第三章 研究方法 27
3.1ANFIS 27
3.1.1 數據來源 27
3.1.2 模糊系統 28
3.1.2.1 模糊推理系統設計 28
3.1.2.2 Takagi-Sugeno-Kang 型系統與 Mamdani 型系統比較 32
3.1.3 適應性模糊類神經推論系統 33
3.1.4 適應性模糊類神經推論系統基本架構 34
3.1.5 歸屬函數定義 43
3.1.6 模式架構之評析 47
3.2 ANAMMOX 50
3.2.1 研究設備 50
2.2.2研究流程 52
2.2.2.1 檢驗方法 53
第四章 結果與討論 55
4.1 ANFIS 55
4.1.1模型之參數篩選 55
4.1.1.1工業區污水處理廠之相關系數結果分析 55
4.1.1.2 模式輸入層變數之選擇 56
4.1.1.3 初步結果 57
4.1.2模型之操作參數 58
4.1.2.1 ANFIS 模型之操作參數 58
4.1.3 ANFIS之SSeff 模擬結果 59
4.1.4 ANFIS之CODeff 模擬結果 63
4.1.5 ANFIS之pHeff 模擬結果 66
4.2 ANAMMOX 70
4.2.1 EABD-AO模廠未添加無機碳源 70
4.2.2 EABD-AO模廠添加無機碳源NH4Cl、NaNO2、KHCO3 74
4.2.3 ANAMMOX批次筒檢測 79
4.2.4 ANAMMOX計量計算 83
第五章 結論與建議 86
5.1 ANFIS 86
5.1.1結論 86
5.2 ANAMMOX 87
5.2.1結論 87
5.3建議 88
參考文獻 89
表目錄
表 2.1 類神經網路與模糊邏輯的優缺點 21
表 3.1 自適應類神經模糊推論系統各層內容一覽表 40
表 3.2 MAPE之預測範圍 50
表 3.3 R值之預測範圍 51
表 3.4 污水廠進流水質概況 54
表 4.1 工業區污水處理廠SSeff之相關系數結果分析表 58
表 4.2 工業區污水處理廠COD eff之相關系數結果分析表 58
表 4.3 工業區污水處理廠pH eff之相關系數結果分析表 58
表 4.4 ANFIS 之預測參數變化 59
表 4.5 ANFIS之基本架構 60
表 4.6 ANFIS之操作參數 61
表 4.7 ANFIS 模式模擬結果 63
表 4.10 未添加無機碳源各項檢測結果 75
表 4.11 添加NH4Cl、NaNO2、KHCO3各項檢測 78
表 4.12 未添加及添加無機碳各項去除率 81
表 4.13 pH和溫度檢測值 83
表 4.14 DO檢測值 84
表 4.15 氨氮檢測值 84
表 4.16 亞硝酸氮檢測值 84
表 4.17硝酸氮檢測值 85
表 4.18批次筒檢測0~6hr所消耗之濃度 88
表 4.19 ANAMMOX計量計算 88
 
圖目錄
圖 1.1研究架構圖 7
圖 3.1 污水處理流程 29
圖 3.2 污泥處理流程 30
圖 3.3 基本模糊邏輯系統架構圖 32
圖 3.4 Mamdani 型系統架構圖 34
圖 3.5 適應性模糊類神經推論系統架構 37
圖 3.6 適應性模糊類神經推論系統模式建立流程圖 48
圖 3.7 EABD-AO 模型廠 54
圖 3.8 批次試驗筒 55
圖 4.1 ANFIS SSeff (2V1) 模式模擬結果 63
圖 4.2 ANFIS SSeff (3V1) 模式模擬結果 64
圖 4.3 ANFIS SSeff (4V1) 模式模擬結果 64
圖 4.4 ANFIS CODeff (2 V 1) 模式模擬結果 67
圖 4.5 ANFIS CODeff (3 V 1) 模式模擬結果 68
圖 4.6 ANFIS CODeff (4 V 1) 模式模擬結果 68
圖 4.7 ANFIS pHeff (2 V 1) 模式模擬結果 71
圖 4.8 ANFIS pHeff (3 V 1) 模式模擬結果 72
圖 4.9 ANFIS pHeff (4 V 1) 模式模擬結果 72
圖 4.10未添加無機碳源氨氮檢測結果 75
圖 4.11未添加無機碳源亞硝酸氮檢測結果 76
圖 4.12未添加無機碳源硝酸鹽氮檢測結果 76
圖 4.13未添加無機碳源COD檢測結果 77
圖 4.14添加NH4Cl、NaNO2、KHCO3 氨氮檢測結果 79
圖 4.15添加NH4Cl、NaNO2、KHCO3 亞硝酸氮檢測結果 79
圖 4.16添加NH4Cl、NaNO2、KHCO3 正磷酸鹽檢測結果 80
圖 4.17添加NH4Cl、NaNO2、KHCO3 COD檢測結果 80
圖 4.18批次試驗氨氮檢測結果 85
圖 4.19批次試驗亞硝酸氮檢測結果 86
圖 4.20批次試驗硝酸氮檢測結果 86

參考文獻
一、英文文獻
1.APHA, AWWA, WEF. Standard Methods for the Examination of Water and Wastewater. 19th ed(1995).
2.Cirpus Irina E.Y., Geerts W., Hermans John H.M., Op den Camp Huub J.M. and Strous M. Challenging protein purification from ANAMMOX bacteria. International Journal of Biological Macromolecules, 39(1-3), 88-94(2006).
3.Dong Z. and Sun T. A potential new process for improving nitrogen removal in constructed wetlands-Promoting coexistence of partial-nitrification and ANAMMOX. Ecological Engineering, 31(2), 69-78(2007).
4.Egli K., Franger U., Alvarez P.J.J., Siegrist H., Vandermeer J.R. and Zehnder A.J.B. Enrichment and characterization of ANANAMMOX bacterium from a rotating biological contactor treating ammonium-rich leachate. Arch Microbiol, 175, 198-207(2001).
5.Engström P., Dalsgaard T., Hulth S. and Aller R.C. Anaerobic ammonium oxidation by nitrite (ANAMMOX): Implications for N2 production in coastal marine sediments.Geochimica et Cosmochimica Acta, 69(8), 2057-2065(2005).
6.Henze M., Gujer W., Mino T. and van Loosdrecht M.C.M. Activated sludge models: ASM1, ASM2, ASM2d and ASM3.IWA, London (2000).
7.Koch G., Egli K., van der Meer J.R., Siegrist H. Mathematical modelling of autotrophic denitrification in a nitrifying biofilm of a rotating biological contactor.Water Science and Technology, 41(4-5) ,91-198(2000).
8.Love N.G., Rust M.E. and Terlesky K.C. Enrichment and characterization of an anaerobic methyl ethyl ketoxime degrading culture from an anoxic/anaerobic/aerobic activated sludge sequencing batch reactor.Water Science and Technology, 37(4-5),5-98(1998).
9.Mahmut Firat. and Mahmud G¨ung¨or. River flow estimation using adaptive neuro fuzzy inference system. Mathematics and Computers in Simulation, 75(75), 89-96 (2007).
10.Malaspina F., Stante L., Cellamare C.M. and Tilche A. Cheese whey and cheese factory wastewater treatment with a biological anaerobic-aerobic process. Water Science and Technology, 32(12), 59-72(1995).
11.Olsson G. and Newell B. Wastewater Treatment Systems. Modelling.
12.Pai T.Y., Chuang S.H., Ho H.H., Yu L.F., Su H.C., and Hu H.C. Predicting performance of grey and neural network in industrial effluent using online monitoring parameters. Process Biochemistry.
13.Pai T.Y., Tsai Y.P., Lo H.M., Tsai C.H. and Lin C.Y. Grey and neural network prediction of suspended solids and chemical oxygen demand in hospital wastewater treatment plant Effluent. Computers & Chemical Engineering, (2007).
14.Pathak B.K., Kazama F., Saiki Y. and Sumino T. Presence and activity of anammox and denitrification process in low ammonium-fed bioreactors. Bioresource Technology, 98(11), 2201-2206 (2007).
15.Prentice-Hall, Inc. Upper Saddle River, New Jersey (1999).
16.Schmidt I., Sliekers O., Schmid M., Cirpus I., Strous M., Bock E. and Kuenen J.G. erobic and anaerobic ammonia oxidizing bacteria-competitors or natural partners. EMS Microbiology Ecology, 39( 3), 175-181(2002).
17.Strous M., Heijnen J.J., Kunen J.G., Jetten M.S.M. The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidising micro-organisms. Applied Microbiology Biotechnology, 50, 589-596(1998).
18.Strous M., Kuenen J.G. and Jetten M.S.M. Key physiology of anaerobic ammonium oxidation. Applied Environmental Microbiology, 65, 3248-3250(1999).
19.Van de Graaf A.A., de Bruijn P., Robertson L.A., Jetten M.S.M. and Kuenen J.G. Metabolic pathway of anaerobic ammonium oxidation on the basis of N-15 studies in a fluidized bed reactor. Microbiology, 143, 2415-2421(1997).
20.Wang J.L. and Kang J. The characteristics of anaerobic ammonium oxidation (ANAMMOX) by granular sludge from an EGSB reactor. Process Biochemistry, 40(5), 1973-1978(2005).
21.Ye R.W. and Thomas S.M. Microbial nitrogen cycles: physiology. genomics and applications, Current Opinion in Microbiology, 4(3), 307-312(2001).
22.Yeh F.H., Lu Y.H., Li C.L. and Wu M.T. Application of ANFIS for inverse prediction of hole profile in thesquare hole bore-expanding process. Materials Processing Technology, 173(2), 136-144 (2006).
23.Yen J., Langari R., Fuzzy logic: intelligence, control, and information.
24.Ying H., Ding Y., Li S. and Shao S. Comparison of Necessary Conditions for Typical Takagi–Sugeno and Mamdani Fuzzy Systems as Universal Approximators. IEEE Transactions on Systems, Man and Cybernetics—PART A: systems and humans, 29(5), 508-514(1999).
25.Yoon T.I., Lee H.S. and Kim C.G. Comparison of pilot scale performances between membrane bioreactor and hybrid conventional wastewater treatment systems. Journal of Membrane Science, 242(1-2), 5-12(2004).
26.Yu X., Qi Z., Zhang X., Yu P., Liu B., Zhang L. and Fu L. Nitrogen loss and oxygen paradox in full-scale biofiltration for drinking water treatment. Water Research, 41(7), 1455-1464(2007).
27.Chang S.C., Pai T.Y., Ho H.H., Leu H.G. and Shieh Y.R. Evaluating Taiwan’s air quality variation trends using grey system theory. Journal of the Chinese Institute of Engineers, 30(2), 361-367 (2007).
28.Pai T.Y., Ouyang C.F., Su J.L. and Leu H.G. Modelling the steady-state effluent characteristics of the TNCU process with ASM2d under varied SRT conditions. Journal of the Chinese Institute of Environmental Engineering, 10(1), 35-42 (2000a).
29.Pai T.Y., Ouyang C.F., Liao Y.C. and Leu H.G. Oxygen transfer in gravity flow sewer. Water Science and Technology, 42(3-4), 417-422 (2000b).
30.Pai T.Y., Ouyang C.F., Su J.L. and Leu H.G. Modeling the stable effluent qualities of the A2O process with Activated Sludge Model 2d under different return supernatant. Journal of the Chinese Institute of Engineers, 24(1), 75-84 (2001a).
31.Pai T.Y., Ouyang C.F., Su J.L. and Leu H.G. Modelling the steady-state effluent characteristics of the TNCU process under different return mixed liquid. Applied Mathematical Modelling, 25(12), 1025-1038 (2001b).
32.Pai T.Y., Chuang S.H., Tsai Y.P. and Leu H.G. Development of two-stage nitrification/denitrification model (TaiWan Extension Activated sludge model NO.1) for BNR process. Journal of the Chinese Institute of Environmental Engineering, 14(1), 51-60 (2004a).
33.Pai T.Y., Tsai Y.P., Chou Y.J., Chang H.Y., Leu H.G. and Ouyang C.F. Microbial kinetic analysis of three different types of EBNR process. Chemosphere, 55(1), 109-118 (2004b).
34.Pai T.Y., Chuang S.H., Tsai Y.P. and Ouyang C.F. Modelling a combined anaerobic/anoxic oxide and rotating biological contactors process under dissolved oxygen variation by using an activated sludge - biofilm hybrid model. Journal of Environmental Engineering-ASCE, 130(12), 1433-1441 (2004c).
35.Pai T.Y. Modeling nitrite and nitrate variations in A2O process under different return oxic mixed liquid using an extended model. Process Biochemistry, 42(6), 978-987 (2007).
36.Pai T.Y., Hanaki K., Ho H.H. and Hsieh C.M. Using grey system theory to evaluate transportation on air quality trends in Japan, Transportation Research Part D: Transport and Environment, 12 (3), 158-166 (2007a).
37.Pai T.Y., Tsai Y.P., Lo H.M., Tsai C.H. and Lin C.Y. Grey and neural network prediction of suspended solids and chemical oxygen demand in hospital wastewater treatment plant effluent, Computers & Chemical Engineering, 31(10), 1272-1281 (2007b).
38.Pai T.Y. Gray and neural network prediction of effluent from the wastewater treatment plant of industrial park using influent quality. Environmental Engineering Science, 25(5), 757-766 (2008).
39.Pai T.Y., Chuang S.H., Ho H.H., Yu L.F., Su H.C. and Hu H.C. Predicting performance of grey and neural network in industrial effluent using online monitoring parameters. Process Biochemistry, 43(2), 199-205 (2008a).
40.Pai T.Y., Chuang S.H., Wan T.J., Lo H.M., Tsai Y.P., Su H.C., Yu L.F., Hu H.C. and Sung P.J. Comparisons of grey and neural network prediction of industrial park wastewater effluent using influent quality and online monitoring parameters. Environmental Monitoring and Assessment, 146(1-3), 51-66 (2008b).
41.Pai T.Y., Chiou R.J. and Wen H.H. Evaluating impact level of different factors in environmental impact assessment for incinerator plants using GM (1, N) model. Waste Management, 28(10), 1915-1922 (2008c).
42.Pai T.Y., Wang S.C., Lo H.M., Chiang C.F., Liu M.H., Chiou R.J., Chen W.Y., Hung P. S., Liao W.C., Leu H.G. Novel modeling concept for evaluating the effects of cadmium and copper on heterotrophic growth and lysis rates in activated sludge process. Journal of Hazardous Materials, 166(1), 200-206 (2009a).
43.Pai T.Y., Wan T.J., Hsu S.T., Chang T.C., Tsai Y.P., Lin C.Y., Su H.C. and Yu L.F. Using fuzzy inference system to improve neural network for predicting hospital wastewater treatment plant effluent. Computers & Chemical Engineering, 33(7), 1272-1278 (2009b).
44.Pai T.Y., Wang S.C., Chiang C.F., Su H.C., Yu L.F., Sung P.J., Lin C.Y. and Hu H.C. Improving neural network prediction of effluent from biological wastewater treatment plant of industrial park using fuzzy learning approach. Bioprocess and Biosystems Engineering, 32(6), 781-790 (2009c).
45.Pai T.Y., Chang H.Y., Wan T.J., Chuang S.H. and Tsai Y.P. Using an extended activated sludge model to simulate nitrite and nitrate variations in TNCU2 process. Applied Mathematical Modelling, 33(11), 4259-4268 (2009d).
46.Pai T.Y., Wang S.C., Lin C.Y., Liao W.C., Chu H.H., Lin T.S., Liu C.C. and Lin S.W. Two types of organophosphate pesticides and their combined effects on heterotrophic growth rates in activated sludge process. Journal of Chemical Technology and Biotechnology, 84(12), 1773-1779 (2009e).
47.Pai T.Y., Wan T.J., Tsai Y.P., Tzeng C.J., Chu H.H., Tsai Y.S. and Lin C.Y. Effect of sludge retention time on biomass and kinetic parameter of two nitrifying species in anaerobic/oxic process. CLEAN-Soil Air Water, 38(2), 167-172 (2010a).
48.Pai T.Y., Chiou R.J., Tzeng C.J., Lin T.S., Yeh S.C., Sung P.J., Tseng C.H., Tsai C.H., Tsai Y.S., Hsu W.J. and Wei Y.L. Variation of biomass and kinetic parameter for nitrifying species in TNCU3 process at different aerobic hydraulic retention time. World Journal of Microbiology & Biotechnology, 26(4), 589-597 (2010b).
49.Pai T.Y., Huang J.D., Wang S.C., Chang D.H., Huang K.J., Lee C.C., Lin S.R., Tseng C.H., Sung P.J. and Leu H.G. Evaluate the establishment site of ecological water purification processes in Dali River using QUAL2K. Suatainable Environment Research, 20(4), 239-243 (2010c).
50.Pai T.Y., Chen C.L., Chung H., Ho H.H. and Shiu T.W. Monitoring and assessing variation of sewage quality and microbial functional groups in a trunk sewer line. Environmental Monitoring and Assessment, 171(1-4), 551-560 (2010d).
51.Pai T.Y., Lin K.L., Shie J.L., Chang T.C. and Chen B.Y. Predicting the co-melting temperatures of municipal solid waste incinerator fly ash and sewage sludge ash using grey model and neural network. Waste Management & Research, 29(3), 284-293 (2011a).
52.Pai T.Y., Ho C.L., Chen S.W., Lo H.M., Sung P.J., Lin S.W., Lai W.J., Tseng S.C., Ciou S.P., Kuo J.L. and Kao J.T. Using seven types of GM (1, 1) model to forecast hourly particulate matter concentration in Banciao City of Taiwan. Water, Air, and Soil Pollution, 217(1-4), 25-33 (2011b).
53.Pai T.Y., Yang P.Y., Wang S.C., Lo H.M., Chiang C.F., Kuo J.L., Chu H.H., Su H.C., Yu L.F., Hu H.C. and Chang Y.H. Predicting effluent from the wastewater treatment plant of industrial park based on fuzzy network and influent quality. Applied Mathematical Modelling, 35(8), 3674-3684 (2011c).
54.Pai T.Y., Hanaki K., Su H.C. and Yu L.F. A 24-h forecast of oxidant concentration in Tokyo using neural network and fuzzy learning approach. CLEAN-Soil Air Water, 2012. (In press)
55.Pai T.Y., Hanaki K. and Chiou R.J. Forecasting hourly roadside particulate matter in Taipei County of Taiwan based on first-order and one-variable grey model. CLEAN-Soil Air Water, 2012. (In press)

 
二、中文文獻
1.余永權，「單片模糊邏輯控制」，北京航空航天大學出版社，北京，大陸(1995)。
2.林基興，「模糊科技的理論與應用」，機電整合雜誌，第19 期，第92-98 頁 (2000)。
3.陳正斌，「應用模糊理論於颱風降雨量之推估」，國立成功大學水力與海洋工程研究所碩士論文，臺南(2004)。
4.陳建谷，「應用倒傳遞類神經及適應性模糊類神經網路模式預測垃圾焚化廠煙道氣之比較研究」，國立雲林科技大學環境與安全工程系碩士論文，雲林(2003)。
5.陶雨龍，「改良式模糊類神經網路與其應用」，私立中原大學工業工程系碩士論文，桃園(2004)。
6.楊國珍，「類神經模糊系統應用於蒸發量推估之研究」，國立成功大學水力與海洋工程研究所碩士論文，臺南(2004)。
7.劉中河，「類神經模糊系統應用於白河水庫防洪放流之研究」，國立成功大學水力與海洋工程研究所碩士論文，臺南(2004)。
8.歐陽嶠暉，「下水道工程學」，長松文化公司，台北(2000)。
9.蔡獲光，「應用 ANFIS 模式於東部海域之季節風波浪推算」，國立交通大學土木工程研究所碩士論文，新竹(2006)。
10.蘇漢昌，「應用適應性模糊推論系統改善類神經網路預測工業廢水廠出流水水質之研究應用」，朝陽科技大學環境工程與管理系碩士論文，台中(2008)。