Aktif çamur prosesi havalandırma havuzu askıda katı madde (AKM) konsantrasyonunun mekanistik, yapay sinir ağı ve hibrit yöntemlerle modellenmesi

Dinamik simülasyon atıksu arıtma tesislerinde işletmenin iyileştirilmesinde önemli bir araçtır. Bu çalışmada, Ankara Merkezi Atıksu Arıtma Tesisinin dinamik simülasyon modeli tasarlanmıştır. Öncelikle, evsel atıksu arıtma prosesinin mekanistik modeli Activated Sludge Model No. 1 bazında GPS-X bilgisayar programı kullanılarak geliştirilmiştir. Yapay Sinir ağı modeli de geriye yayılım algoritmasını esas alan MLP sinir ağı yardımı ile oluşturulmuştur. Daha sonra, mekanistik model yapay sinir ağı ile birleştirilmiştir. Yapay sinir ağı modellerinin en uygun ağ yapısı modellerin birçok adımda eğitilmesi ve test edilmesi ile tespit edilmiştir. Her üç model, prosesin dinamik davranışını tahmin etmek için tesisinin işletme ve laboratuar analizlerinden elde edilen aynı veriler ile oluşturulmuştur. Havalandırma tankı Askıda Katı Madde (AKM) konsantrasyonu tahmin edilmiş ve sonuçları karşılaştırılmıştır. Hibrit model yaklaşımının daha başarılı sonuçlar verdiği ve tesisin işletme koşullarının ASM1 ve YSA modellerinden daha iyi tanımlandığı gözlenmiştir.

Modeling of activated sludge process of aeration tank mixed liquor suspended solids (MLSS) concentrations by using mechanistic, Artifical neural network and hybrid model approaches

Dynamic simulation is an important tool for the improvement of wastewater treatment plant operation. In this study, dynamic simulation model of the Ankara central wastewater treatment plant (ACWT) were evaluated. First, a mechanistic model of the municipal wastewater treatment process is developed based on Activated Sludge Model No. 1 by using a GPS-X computer program. Artificial neural network model is also developed with the help of MLP neuronal networks based on back-propagation algorithm. Then, the mechanistic model is combined with artificial neural network in parallel configuration. The appropriate architecture of the neural network models was determined through several steps of trainings and testing of the models. Both three models are performed with the same data obtained from the plant operation and laboratory analysis to, predict dynamic behaviour of the process. Using these three models, by the purpose of evaluation of treatment performance, aeration tank MLSS concentrations have been predicted and the results have been compared. It is observed that the hybrid model approach gives more successful results and describes the operation conditions of the plant better than ASM1 and ANN.

___

Chen, J.C., Chang, N.B. and Shieh, W.K. 2003. Assessing wastewater reclamation potential by neural network model. Engineering Applications of Artificial Intelligence 16: 149-157.
Cote, Mv Grandjean, B.P.A., Lessard, P., Thibault, J. 1995. Dynamic modeling of the activated sludge process: improving prediction using neural networks. Water Research 29(4): 995-1004.
Gokcay, C.F. and Sin, G. 2004. Modelling of a large-scale wastewater treatment plant for efficient operation. Water Science and Technology 50(7): 123-130.
GPS-X version 5.0 user's guide, 2006. Canada: JT Hydromantis Inc.
Gujer, W., Henze, M., Mino, T.,& van Loosdrechi, M. 1999. Activated sludge model no. 3. Water Science Technology 39(1): 183-193.
Güçlü, D., 2007. Tam Ölçekli Kentsel Atıksu Arıtma Tesislerinin Bilgisayar Programı kullanılarak Modellenmesi ve Arıtma Performanslarının İncelenmesi. Doktora Tezi, S.Ü. Fen Bilimleri Enstitüsü, 2007, Konya.
Hamed, M.M., Khalafallah, M.G. and Hassanien, E.A. 2004. Prediction of wastewater treatment plant performance using artificial neural networks. Environmental Modelling & Software 19: 919-928.
Hack, M. and Köhne, M. 1996. Estimation of wastewater process parameters using artificial neural networks. Water Science and Technology 33(1): 101-115.
Henze, M., Grady C. P. L., Jr, Gujer, W., Marais, G. v. R., Matsuo, T. 1987. Activated sludge model no 1. IAWQ Scientific and Technical Report No 1, London, UK.
Henze, M., Gujer, W., Mino, T., Matsuo, Tv Wentzel, M. C, Marais, G. v. R., Van Loosdrecht, M. C. M. 1995. Activated sludge model no 2. IAWQ, Scientific and Technical Report No 3, London, UK.
Ladiges G., Gunner, C. und Otterpohl, R. 2001. Optimierung des Hamburger Klarwerksverbundes Köhlbrandhöft/Dradenau mithilfe der dynamischen Simulation. KA-Wasserwirtschaft, Abwasser, Abfall 48(4): 490-498.
Lee, D. S., Vanrolleghem, P.A., Park, J.M. 2005. Parallel hybrid modeling methods for a full-scale cokes wastewater treatment plant. Journal of Biotechnology 115: 317-328.
Morgenroth, E., Arvin, E., Vanrolleghem, P. 2002. The use of mathematical Models in teaching Wastewater treatment engineering. Water Science and Technology 45(6): 229-233.
Nuhoglu, A., Keskinler, B., Yildiz, E. 2005. Mathematical modelling of the activated sludge process-the Erzincan case. Process Biochemistry 40: 2467-2473.
Onkal-Engin, G., Demir, I. and Engin, S.N. 2005. Determination of the relationship between sewage odour and BOD by neural networks. Environmental Modelling & Software 20: 843-850.
Winkler, U. und Voigtlânder, G. 1995. Anwendung neuronaler Netze für die Simulation von Prozessablâufen auf vorhandenen Klâranlagen. Korrespondenz Abwasser 10:1784-1792.
Zhao, H., Hao, O.J. and McAvoy, T.J. 1999. Approaches to modeling nutrient dynamics: ASM2, simplified model and neural nets. Water Science and Technology 39(1): 227-234.