BİYOLOJİK ATIKSU ARITMA TESİSLERİNDE KARBON VE AZOT GİDERİM VERİMLERİNİN TAHMİNİ AMACIYLA YAPAY SİNİR AĞLARININ KULLANIMI

Biyolojik azot giderimi gerçekleştirilen atıksu arıtma proseslerinin (AO prosesi) modellenmesi amacıyla Aktif Çamur Modeli No. 1 (ASM1) kullanılagelmişse de bu modelde ihtiyaç duyulan girdi parametrelerinin tahmin edilmesi çok zaman almaktadır. Bu çalışma kapsamında, ASM1 kadar detaylı girdi verisi gerektirmeyen geri beslemeli yapay sinir ağlarının (BPANN) AO proseslerindeki kimyasal oksijen ihtiyacı (KOİ), toplam Kjeldahl azotu (TKN) ve toplam azot (TN) giderim verimlerinin tahminindeki performansı test edilmiştir. Bu amaçla BPANN’de dört farklı aktivasyon fonksiyonu kullanılmıştır. Elde edilen sonuçlar, AO proseslerindeki KOİ, TKN ve TN giderim verimlerinin BPANN ile yüksek doğrulukta tahmin edilebildiğini göstermiş; en iyi öğrenme ve tahmin yeteneği ise Sinc fonksiyonu ile elde edilmiştir. Sinc-BPANN ile elde edilen ortalama kare hatalar KOİ giderim verimi için 2,50×10-4, TKN giderim verimi için 4,15×10-4, TN giderim verimi için ise 2,65×10-4 olarak hesaplanmıştır. Buna göre Sinc-BPANN AO proseslerindeki KOİ, TKN ve TN giderim verimlerinin doğrusal olmayan doğasını ASM1’e nazaran çok daha az girdi parametresiyle açıklayabilmektedir.

Anahtar Kelimeler:

ASM1, yapay sinir ağları, karbon giderimi, azot giderimi, aktivasyon fonksiyonu

USE OF ARTIFICIAL NEURAL NETWORKS AS A TOOL TO PREDICT CARBON AND NITROGEN REMOVAL EFFICIENCIES IN BIOLOGICAL WASTEWATER TREATMENT PLANTS

Although Activated Sludge Model No. 1 (ASM1) was used for modelling biological nitrogen removal processes, estimation of input parameters required to run this model necessitates complicated laboratory analyses. In this study, the performance of Backpropagation Artificial Neural Networks (BPANN), which requires considerably less numbers of input parameters, in predicting chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN), and total nitrogen (TN) removal efficiencies was tested. For this purpose, four activation functions were employed in BPANN. Results suggested that COD, TKN, and TN removal efficiencies in AO processes can be accurately estimated using BPANN, with the highest learning and prediction capacity when Sinc function is employed. The mean square errors (MSEs) with Sinc-BPANN were calculated as 2.50×10-4 for COD removal efficiency, 4.15×10-4 for TKN removal efficiency, and 2.65×10-4 for TN removal efficiency. Therefore, the Sinc-BPANN is concluded to be an efficient tool for estimating nonlinear nature of COD, TKN, and TN removal efficiencies in AO processes using considerably less numbers of input parameters.

Keywords:

ASM1, Artificial Neural Networks, Carbon Removal, Nitrogen Removal, activation function,

PDF

___

[1] https://esa.un.org/unpd/wpp/Download/Standard/Population/# (erişim tarihi 15.12.2016).
[2] MANAV DEMİR, N., “İleri Biyolojik Arıtma Proseslerinde Nütrient Giderimi ve Mikroorganizma Türlerinin İncelenmesi”, Doktora Tezi, Yıldız Teknik Üniversitesi Fen Bilimleri Enstitüsü, İstanbul, Türkiye, 2012.
[3] HENZE, M., GRADY, C.P.L., GUJER, W., MARAIS, G.V.R., MATSUO, T., Activated Sludge Model No. 1. In M. HENZE, W. GUJER, T. MINO, M.V. LOOSDRECHT (Eds.), Activated Sludge Models ASM1, ASM2, ASM2d and ASM3 (pp. 1-37), IWA Publishing, London, UK, 2007.
[4] HENZE, M., GUJER, W., MINO, T., MATSUO, T., WENTZEL, M.C., MARAIS, G.V.R., Activated Sludge Model No. 2. In M. HENZE, W. GUJER, T. MINO, M.V. LOOSDRECHT (Eds.), Activated Sludge Models ASM1, ASM2, ASM2d and ASM3 (pp. 39-73), IWA Publishing, London, UK, 2007.
[5] HENZE, M., GUJER, W., MINO, T., MATSUO, T., WENTZEL, M.C., MARAIS, G.V.R., “Activated Sludge Model No. 2d”, Water Science and Technology, 39, 165-182, 1999.
[6] GUJER, W., HENZE, M., MINO, T., VAN LOOSDRECHT, M.C.M., “Activated Sludge Model No. 3”, Water Science and Technology, 39, 183-193, 1999.
[7] HUG, T., BENEDETTI, L., HALL, E.R., JOHNSON, B.R., MORGENROTH, E., NOPENS, I., RIEGER, L., SHAW, A., VANROLLEGHEM, P.A., “Wastewater Treatment Models in Teaching and Training: The Mismatch between Education and Requirements for Jobs”, Water Science and Technology, 60, 1721-1729, 2009.
[8] COPP, J.B., MURPHY, K.L., “Estimation of the Active Nitrifying Biomass in Activated Sludge”, Water Research, 29, 1855-1862, 1995.
[9] SMETS, I.Y., HAEGEBART, J.V., CARETTE, R., VAN IMPE, J.F., “Linearization of the Activated Sludge Model ASM1 for Fast and Reliable Predictions”, Water Research, 37, 1831-1851, 2003.
[10] MULLER, A., WENTZEL, M.C., LOEWENTHAL, R.E., EKAMA, G.A., “Heterotroph Anoxic Yield in Anoxic Aerobic Activated Sludge Systems Treating Municipal Wastewater”, Water Research, 37, 2435-2441, 2003.
[11] SONG, Y., XIE, Y., YUDIANTO, D., “Extended Activated Sludge Model No. 1 (ASM1) for Simulating Biodegradation Process Using Bacterial Technology”, Water Science and Engineering, 5, 278-290, 2012.
[12] JANUS, T., ULANICKI, B., “ASM1-Based Activated Sludge Model with Polymer Kinetics for Integrated Simulation of Membrane Bioreactors for Wastewater Treatment”, Procedia Engineering, 119, 1318-1327, 2015.
[13] CRUZ, J.A.S., MUSATTI, S.F., SCENNA, N.J., GERNAEY, K.V., MUSATTI, C.M., “Reaction Invariant-Based Reduction of the Activated Sludge Model ASM1 for Batch Applications”, Journal of Environmental Chemical Engineering, 4, 3654-3664, 2016.
[14] ZHAO, B., SU, Y., “Artificial Neural Network-Based Modeling of Pressure Drop Coefficient for Cyclone Separators”, Chemical Engineering Research and Design, 88, 606-613, 2010.
[15] AGHAV, R.M., KUMAR, S., MUKHERJEE, S.N., “Artificial Neural Network Modeling in Competitive Adsorption of Phenol and Resorcinol from Water Environment Using Some Carbonaceous Adsorbents”, Journal of Hazardous Materials, 188, 67-77, 2011.
[16] HERNANDEZ-RAMIREZ, D.A., HERRERA-LOPEZ, E.J., RIVERA, A.L., DEL REAL-OLVERA, J., “Artificial Neural Network Modeling of Slaughterhouse Wastewater Removal of COD and TSS by Electrocoagulation”, Studies in Fuzziness and Soft Computing, 312, 273-280, 2014.
[17] SAMLI, R., SIVRI, N., SEVGEN, S., KIREMITCI, V.Z., “Applying Artificial Neural Networks for the Estimation of Chlorophyll-a Concentrations along the Istanbul Coast”, Polish Journal of Environmental Studies, 23, 1281-1287, 2014.
[18] BAYRAM, A., KANKAL, M., “Artificial Neural Network Modeling of Dissolved Oxygen Concentrations in a Turkish Watershed”, Polish Journal of Environmental Studies, 24, 1507-1515, 2015.
[19] DEMİR, S., KARADENİZ, A., MANAV DEMİR, N., “Using Steepness Coefficient to Improve Artificial Neural Network Performance for Environmental Modeling”, Polish Journal of Environmental Studies, 25, 1467-1477, 2016.
[20] DEMİR, S., KARADENİZ, A., MANAV DEMİR, N., “Artificial Neural Network Simulation of Cyclone Pressure Drop: Selection of the Best Activation Function”, Polish Journal of Environmental Studies, 25, 1891-1899, 2016.
[21] HAMED, M.M., KHALAFALLAH, M.G., HASSANIEN, E.A., “Prediction of Wastewater Treatment Plant Performance Using Artificial Neural Networks”, Environmental Modeling and Software, 19, 919-928, 2004.
[22] MJALLI, F., AL-ASHEH, S., ALFADALA, H.E., “Use of Artificial Neural Network Black-Box Modeling for the Prediction of Waterwater Treatment Plants Performance”, Journal of Environmental Management, 83, 329-338, 2007.
[23] MORAL, H., AKSOY, A., GOKCAY, C.F., “Modeling of the Activated Sludge Process by Using Artificial Neural Newtworks with Automated Architecture”, Computers and Chemical Engineering, 32, 2471-2478, 2008.
[24] TAKACS, I., PATRY, G.G., NOLASCO, D., “A Dynamic Model of the Clarification-Thickening Process”, Water Research, 25, 1263-1271, 1991.
[25] TAKACS, I., “Experiments in Activated Sludge Modeling”, PhD Thesis, Ghent University Applied Biological Sciences, Belgium, 2008.
[26] HOLENDA, B., PASZTOR, I., KARPATI, A., REDEY, A., “Comparison of One Dimensional Secondary Settling Tank Models”, E-Water, 06, 1-17, 2006.