Atık Su Miktarının ARIMA ve Yapay Sinir Ağları ile Tahmini

Yıl 2025, Cilt: 25 Sayı: 2, 359 - 368

Ayşegül Yıldız , Sermin Elevli , Mehmet Serhat Odabas

Öz

Atık su akış tahmini, atık su arıtma tesislerinin doğru ve etkin bir şekilde yönetimi için anahtar rol oynamaktadır. Kontrolsüz şehirleşme, nüfus artışları, iklim değişikliğinden kaynaklı aşırı yağışlar ve altyapı yetersizlikleri gibi nedenlerden kaynaklanan tutarsız veri ve belirsizlikler atık su akış tahminini güçleştirmektedir. Bu kapsamda uzun vadeli eğilimleri kapsayacak etkili tahmin modellerinin kullanılması ihtiyacı belirgin hale gelmiştir. Bu çalışmada Samsun’un Doğu İleri Biyolojik Atık Su Arıtma Tesisi için atık su akış miktarının bir zaman serisi analiz modeli olan ARIMA ve yapay sinir ağları ile tahmin edilmesi amaçlanmıştır. Bir yıllık süreye karşılık gelen günlük akış miktarı verileri kullanılan çalışmada modellerin performansları RMSE, MAE ve MAPE değerleri açısından karşılaştırılmıştır. ARIMA (2, 1, 2) modeli daha yüksek doğrulukta performans göstermiştir.

Anahtar Kelimeler

ARIMA, Atıksu, Öngörü, Yapay Sinir Ağları

Etik Beyan

etik kurul beyanı gerekmemektedir.

Destekleyen Kurum

Destekleyen kurum yok

Estimation of Wastewater Amount with ARIMA and Artificial Neural Networks

Öz

Wastewater flow estimation plays a key role for the accurate and efficient management of wastewater treatment plants. Inconsistent data and uncertainties arising from uncontrolled urbanization, population increases, excessive rainfall due to climate change and infrastructure deficiencies make wastewater flow forecasting difficult. In this context, the need to use effective forecasting models that will cover long-term trends has become evident. In this study, it is aimed to estimate the amount of wastewater flow for Samsun's East Advanced Biological Wastewater Treatment Plant with ARIMA, a time series analysis model, and artificial neural networks. Daily flow rate data corresponding to a period of one year were used and the performances of the models were compared in terms of RMSE, MAE and MAPE values. ARIMA (2, 1, 2) model showed higher accuracy.

Anahtar Kelimeler

ARIMA, Wastewater, Forecasting, Artificial Neural Networks

Kaynakça

• Al-Dahidi, S., Alrbai, M., Al-Ghussain, L., Alahmer, A. and Hayajneh, H.S., 2024. Data-driven analysis and prediction of wastewater treatment plant performance: Insights and forecasting for sustainable operations. Bioresource Technology, 391, 129937. https://doi.org/10.1016/j.biortech.2023.129937
• AL-Zubaidi, E.D.A., Yas, A.H. and Abbas, H.F., 2019. Guess the time of implementation of residential construction projects using neural networks ANN. Periodicals of Engineering and Natural Sciences, 7(3): 1218-1227, 2019. https://doi.org/10.21533/pen.v7i3.680
• Ataseven, B., 2013. Forecasting by using artificial neural networks. Öneri Dergisi, 10(39), 101-115. https://doi.org/10.14783/od.v10i39.1012000311
• Baki, O.T. and Aras, E., 2018. Estimation of Bod in wastewater treatment plant with different regression models. Engineering Sciences, 13(2), 96-105.
• Çelik, N., Coşar, D.N. and Konyalıoğlu, A.K., 2022. Wastewater Forecasting Application by an Integrated Interpolation and Box-Jenkins Modelling Approach in Turkey. The International Symposium for Production Research, Springer. https://doi.org/10.1007/978-3-031-24457-5_10
• Do, P.C., Chow, W.K., Rameezdeen, R. and Gorjian, N., 2022. Wastewater inflow time series forecasting at low temporal resolution using SARIMA model: a case study in South Australia. Environmental Science and Pollution Research, 29(47), 70984-70999 https://doi.org/10.1007/s11356-022-20777-y
• Elevli, S., 2020. The utilisation of special cause control charts ın the presence of autocorrelated data. Sigma Journal of Engineering and Natural Sciences, 38(2), 787-793
• Er Aydın, B., Odabas, M:S., Senyer, N. and Ardali, Y. 2022. Evaluation of Deep Sea discharge systems efficiency in the eastern black sea using artificial neural network: a case study for Trabzon, Turkey. Brazilian Archives of Biology and Technology, 65: e22210397 https://doi.org/10.1590/1678-4324-2022210397
• Erden, C., 2023. Performance Comparisons of Deep Learning and ARIMA: A Borsa Istanbul Stock Example. Yönetim ve Ekonomi Dergisi, 30(3), 419-438. https://doi.org/10.18657/yonveek.1208807
• Işık, H.E., Bas, E., Egrioglu, T. and Akkan, A., 2024. A new single multiplicative neuron model artificial neural network based on black hole optimization algorithm: forecasting the amounts of clean water given to metropolis. Stoch Environ Res Risk Assess. https://doi.org/10.1007/s00477-024-02802-3
• Kaya, N.S., Pacci, S., Demirağ Turan, İ., Odabas, M.S. and Dengiz, O., 2023. Comparing geographic information systems-based fuzzy-analytic hierarchical process approach and artificial neural network to characterize soil erosion risk indexes. Rendiconti Lincei. Scienze Fisiche e Naturali, 34(4), 1089-1104. https://doi.org/10.1007/s12210-023-01201-0
• Lai, M.S., Wulff, S., Cao, Y., Robinson, T.J. and Rajapaksha, R., 2023. An interpretable time series machine learning method for varying forecast and nowcast lengths in wastewater-based epidemiology. MethodsX, 11, 102382 https://doi.org/10.1016/j.mex.2023.102382
• Namasudra, S., Dhamodharavadhani, S. and Rathipriya, R., 2021. Nonlinear neural network based forecasting model for predicting COVID-19 cases. Neural Processing Letters, 55, 171-191. https://doi.org/10.1007/s11063-021-10495-w
• Odabas, M.S., Senyer, N., Kayhan, G. and Ergun, E., 2017. Estimation of chlorophyll concentration index at leaves using artificial neural networks. Journal of Circuits, Systems and Computers, 26(2), 1750026. https://doi.org/10.1142/S0218126617500268
• Öztemel, E. and Dügenci, M., 2016. Atıksu arıtma tesis kontrolde yapay sinir ağı ile kirlilik parametre tahmini. 3rd International Symposium on Environment and Morality, Alanya, Türkiye.
• Palabıyık, S. and Akkan, T., 2024. Evaluation of water quality based on artificial intelligence: performance of multilayer perceptron neural networks and multiple linear regression versus water quality indexes. Environment, Development and Sustainability https://doi.org/10.1007/s10668-024-05075-6
• Rahman, A. and Hasan, M.M. 2017. Modeling and forecasting of carbon dioxide emissions in Bangladesh using Autoregressive Integrated Moving Average (ARIMA) models. Open Journal of Statistics, 7(4), 560-566 https://doi.org/10.4236/ojs.2017.74038
• Suhermi, N., Suhartono, D.D. and Prastyo, B.A., 2018. Roll motion prediction using a hybrid deep learning and ARIMA model. Procedia computer science, 144, 251-258 https://doi.org/10.1016/j.procs.2018.10.526
• Zhang, G.P., 2003. Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159-175. https://doi.org/10.1016/S0925-2312(01)00702-0
• Zhang, Q., Li, Z., Snowling, S., Siam, A. and El-Dakhakhni, W. 2019. Predictive models for wastewater flow forecasting based on time series analysis and artificial neural network. Water Science and Technology, 80(2), 243-253 https://doi.org/10.2166/wst.2019.263