| 摘要: | 微生物的生化反應在下水道系統及廢水生物處理程序中相當重要。微生物的反應機制相當複雜，使用數學模式可解釋其動力特性，另外，亦可應用於廢水生物處理程序的規劃、設計、操作、控制及診斷。 本研究的目的為：(1)依據活性污泥模式1 (ASM1) 建立數學模式以描述下水道水質的轉化；(2)依據活性污泥模式2d (ASM2d)及Mixed Culture Biofilm Model (MCBM) 建立修正數學模式以描述A2O，TNCU，TNCU1，AOAO及TNCUO程序中碳氮磷各種成份的轉化；(3)比較各試程實驗值與模式預測值在不同操作狀態下的一致性；(4)以模式分析異營菌(XH)、磷蓄積菌(XPAO)、自營菌(XAUT)等不同微生物族群在不同操作狀態下的動力特性。各種程序之相關水質數據由模廠實驗求得。 在建立下水道水質生化動力模式的研究中，以長21.7 m直徑15 cm的下水道模型探討下水道中溶氧於垂直方向傳輸及水質的變化。依據Fick’s law推導一包括分子擴散項及亂流擴散項的偏微分方程。以非線性最小平方法回歸求得溶氧於垂直方向之水力傳輸函數值。結果顯示，垂直方向之水力傳輸函數值是分子擴散、亂流擴散及流速的函數。在依據ASM1建立數學模式以描述下水道水質轉化的實驗中，模式值與實驗值相當一致，可知底部生物膜是降解污染物的主要貢獻，XH缺氧生長是影響各成份轉化途徑的主要因素；系統中XAUT極少甚至不存在。 在建立活性污泥營養鹽處理程序模式的研究中，模式考慮三項修正：(1)溶解性COD(SS)之生物吸附效應：(2)進流污水中有機氮的水解：(3)厭氧槽中異營菌的生長作用。模型廠進流水水質水量固定，變動操作條件。當系統達到穩態出流時，分別比較五種程序各試程之實驗值與模式值。結果顯示模式值與實驗值相當一致。由此一致性可知，SS之生物吸附效應及進流水中有機氮水解是活性污泥系統中相當重要的效應。其次，可知XH於厭氧槽中進行生長。更進一步間接證明活性污泥營養鹽處理程序內存在脫硝除磷菌。 在A2O，TNCU，及TNCU1程序中，厭氧槽中XH，XPAO及XAUT 的濃度會減少，因為裂解反應是厭氧槽的主要反應。好氧槽中，XH，XPAO及XAUT進行好氧生長而增加。磷蓄積菌體內貯存聚磷酸鹽(XPP)在厭氧槽中會減少，後段好氧槽則增加，最後維持一穩定值。磷蓄積菌體內有機碳貯存物(XPHA)在厭氧槽中會增加而在後段好氧槽中被消耗。最後，XPHA維持一穩定值。另外，TNCU1程序之生物膜行為以ASM2d及MCBM進行模擬，.模擬結果顯示：好氧段溶氧越高，生物膜中XAUT之比率及硝酸氮越高，XI之比率及氨氮降低；XH之比率及SS未明顯隨溶氧變動而改變。 TNCUO及AOAO程序可藉段數的重新配置及分段進流的策略有效去除氮、磷，結果顯示：(1)碳、總磷、總氮的有效去除率分別為86.8~97.6, 92~100及62.9~76.4 %：(2)TNCUO及AOAO程序中微生物的動力特性受到不同操作條件的影響：(3)在前段好氧槽中，因HRT較長及可利用基質較多，故XH，XPAO及XAUT行好氧生長而增加。在分流流入的缺氧槽中，因進流的稀釋作用，XH大量減少；但在有分流流入的缺氧槽中，因供給外部碳源，故脫硝持續進行：(4)在未循環硝化混合液及未於缺氧段添加外部碳源以供脫硝的情況下，TNCUO及AOAO程序亦可有效操作。 本研究結果，實驗值與模式預測值相當一致，模式可應用於實廠操作。 The microbial biochemical reactions were very important in the sewer systems and biological wastewater treatment processes. Because the microbial mechanisms were very complex, the mathematical models could be used to explain the microbial kinetic behaviors. Additionally, the mathematical model could be utilized on the planning, design, operation, control and diagnosis of the biological wastewater treatment processes too. The objectives of this study are listed as follows: (1) to establish a mathematical model based on the kinetic of Activated Sludge Model No.1(ASM1) to describe the water quality variations in the sewer systems, (2) to establish a modified mathematical model based on the kinetic of Activated Sludge Model No.2d (ASM2d) and Mixed Culture Biofilm Model (MCBM) to describe the transformation of different compounds including carbon, nitrogen, and phosphorus in the A2O, TNCU, TNCU1, AOAO and TNCUO processes, (3) to explore the consistency between the calculated and observed values of different components by the modified model under different operation conditions in the processes, and (4) to analyze the kinetics of different microorganisms including heterotrophic organisms(XH), phosphorus accumulating organisms(XPAO), and autotrophic organisms(XAUT) in the processes by using the mathematical model. In the study of establishing the sewer water quality model, oxygen transfer and water quality variations in gravity sewer pipes were studied in a 21 m long, 0.15 cm diameter model sewer. A partial differential equation that considered the Fick's law including the molecular and eddy diffusion terms were derived. The analytic solution of the partial differential equation was used to determine the oxygen transfer function values in the vertical direction by the method of nonlinear regression. The oxygen transfer function in the vertical direction is a function of molecular diffusion, eddy diffusion and flow velocity. In the experiments of the water quality variations in sewer pipe, the model values test values showed a good consistency. The results indicated that the biofilm attached on the pipe bottom predominated the degradation of pollutants. The anoxic growth of XH affected the transformation pathways of different components. XAUT did not exist in the system. In the study of establishing the nutrient removal activated sludge model, there were three modifications: (1)the biosorption effect of the soluble COD(SS), (2)the ammonification of the organic nitrogen in influent wastewater, and (3)the XH growth in the anaerobic tank. The influent wastewater qualities were fixed and the operation conditions were varied. When a steady state was reached, the comparisons between the model predicted values and measured values in each test were made in five types of processes. It showed a good consistency between them. According to the consistent results, the biosorption effect of SS and ammonification of the organic nitrogen in the influent wastewater were the important effects in activated sludge system. Additionally, the XH might grow in the anaerobic tank. Furthermore, it was indirectly proved that the denitrifying PAOs existed in the nutrient removal activated sludge system. In A2O, TNCU, and TNCU1 processes, the XH, XPAO, and XAUT would decrease in the anaerobic tank due to the lysis reactions and would increase in the aerobic tank. The polyphosphate (XPP) would decrease in the anaerobic tank and increase in the consequent aerobic tank due to the recovery of XPP level. At last, XPP revealed stable values under different operation conditions. The polyhydroxy-alkanoates (XPHA) would increase in the anaerobic tank and be consumed in the consequent aerobic tank. The concentration of XPHA maintained stable values eventually. On the other hand, the biofilm kinetic characteristics in TNCU1 process was modeled by using ASM2d and MCBM. The results indicated that when dissolved oxygen concentrations increased in the aerobic tank, the fraction of XAUT and nitrate would increase, the fraction of XI and ammonia nitrogen would decrease in the biofilm. The fraction of XH and SS in biofilm would not be affected when dissolved oxygen varied in the aerobic tank. The nitrogen and phosphorus were removed efficiently in the TNCUO and AOAO processes by compartment reconfiguration and stepwise feeding strategy. The results obtained in this study can be summarized as follows: (1)The effective removal efficiency of carbon, T-P and T-N at 86.8~97.6, 92~100 and 62.9~76.4 %, respectively, were achieved in these testing runs. (2)The microbial kinetics would be affected by different operations. (3)When the step feeding strategy was adopted, the HRT was longer due to the less influent flowrate in the front stages and the microbes would grow in quantities in the aerobic reactors. In the followed anoxic reactors, the microbes would decrease in quantities due to the dilution effect. (4)The TNCUO and AOAO processes could be operated more efficiently without additional energy for nitrified liquid circulation and addition of external carbon substrate for denitrification in the anoxic zones. Since the model simulation results showed a good consistency with the test values, the model could be applied on operations of the full scale plant. |
