Aylık Akış Tahmini için Stokastik, Yapay Sinir Ağı ve Dalgacık Bazlı Modellerin Kullanımı

Öz

Yağış ve akış gibi hidrolojik verilerin tahmini için farklı modellerin geliştirilmesi gelecekte su ile ilgili problemlerle mücadele edebilmek açısından önemlidir. Bu çalışma, Yapay Sinir Ağı (ANN), Otoregresif Bütünleşik Hareketli Ortalama (ARIMA), Dalgacık-ARIMA (WARIMA) ve WARIMA-ANN modellerinin aylık akım tahmin performanslarını araştırmaktadır. Bu modeller, Türkiye’nin Susurluk havzasındaki iki istasyonda uygulanmıştır. Bu bağlamda, ilk olarak akış verileri WARIMA ve WARIMA-ANN modelleri için dalgacık dönüşümü ile bileşenlerine ayrılmıştır. Daha sonra, her bir model için akış tahminleri gerçekleştirilmiştir. Karşılaştırma ölçütü olarak, Hataların Ortalama Karakökü (RMSE), Kling-Gupta Verimliliği (KGE) ve Nash Sutcliffe Verimliliği (NSE) göz önünde bulundurulmuştur. Sonuç olarak, WARIMA ve WARIMA-ANN modellerinin, özellikle ARIMA ve ANN modellerine göre daha iyi performans gösterdiği tespit edilmiştir.Buna ek olarak, dalgacık dönüşümünün ARIMA ve ARIMA-ANN modellerinin performansını geliştirdiği belirgin şekilde görülmüştür.

Anahtar Kelimeler

• ANN
• ARIMA
• Akış
• Dalgacık
• Türkiye

Utilization of Stochastic, Artificial Neural Network, and Wavelet Combined Models for Monthly Streamflow

Öz

The development of various models to estimate hydrological variables, such as precipitation and runoff is significant regarding handling the water-related problems in the future. This study investigates the performances of Artificial Neural Network (ANN), Auto-Regressive Integrated Moving Average (ARIMA), Wavelet-ARIMA (WARIMA), and WARIMA-ANN models for monthly streamflow forecasting. These models were utilized in two stations of the Susurluk basin in Turkey. In this regard, first, the streamflow data were decomposed into components by wavelet transformation for the WARIMA and WARIMA-ANN models. After that, runoff predictions were performed for each model. As comparison criteria, Root Mean Square Error (RMSE), Kling Gupta Efficiency (KGE), and Nash Sutcliffe Efficiency (NSE) were taken into consideration. As a result, it was obtained that WARIMA and WARIMA-ANN models performed better than the ARIMA and ANN models, particularly. In addition, it was seen that wavelet transformation improved the performance of ARIMA and ARIMA-ANN models, obviously. 

Anahtar Kelimeler

• ANN
• ARIMA
• Streamflow
• Wavelet
• Turkey

Teşekkür

The authors are grateful to General Directorate of State Hydraulic Works for providing the data. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Authors declare that they have no conflict of interest.

Kaynakça

1. Jain, A. & Kumar A. M. (2007). Hybrid neural network models for hydrologic time series forecasting. Applied Soft Computing, 7(2), 585-592.
2. Adamowski, J. & Chan, H. F. (2011). A wavelet neural network conjunction model for groundwater level forecasting. Journal of Hydrology, 407(1-4), 28-40.
3. Wang, W.C, Chau, K.W., Xu, D.M. & Chen X.Y. (2015). Improving forecasting accuracy of annual runoff time series using ARIMA based on EEMD decomposition. Water Resources Management, 29(8), 2655-2675.
4. Aqil, M, Kita, I., Yano, A. & Nishiyama, S. (2007). Neural networks for real time catchment flow modeling and prediction. Water Resources Management, 21(10), 1781-1796.
5. Lin, G. F., Chen, G. R., Huang, P. Y., & Chou, Y. C. (2009). Support vector machine-based models for hourly reservoir inflow forecasting during typhoon-warning periods. Journal of Hydrology, 372(1-4), 17-29.
6. Zadeh, M. R., Amin, S., Khalili, D. & Singh, V.P. (2010). Daily outflow prediction by multilayer perceptron with logistic sigmoid and tangent sigmoid activation functions. Water resources management, 24(11), 2673-2688.
7. Kurtuluş, B. & Razack, M. (2010). Modeling daily discharge responses of a large karstic aquifer using soft computing methods: artificial neural network and neuro-fuzzy. Journal of Hydrology, 381(1-2), 101-111.
8. Goyal, M.K., Sharma, A., Katsifarakis, K.L. (2017). Prediction of flow rate of karstic springs using support vector machines. Hydrological Sciences Journal, 62(13), 2175-2186.

9. Shafaei, M., Adamowski, J., Fakheri-Fard, A., Dinpashoh, Y. & Adamowski, K. (2016). A wavelet-SARIMA-ANN hybrid model for precipitation forecasting. Journal of Water and Land Development, 28(1), 27-36.
10. Kaur, D., Lie, T. T., Nair, N. K. & Vallès, B. (2015) Wind speed forecasting using hybrid wavelet transform-ARMA techniques. AIMS Energy, 3(1), 13-24.
11. Valipour, M., Banihabib, M. E. & Behbahani, S. M. R. (2013). Comparison of the ARMA, ARIMA, and the autoregressive artificial neural network models in forecasting the monthly inflow of Dez dam reservoir, Journal of Hydrology, 476, 433-441.
12. Pektas, A. O. & Cigizoglu, H.K, (2017). Long-range forecasting of suspended sediment, Hydrological Sciences Journal, 62(14), 2415-2425.
13. Liu, Y., Wu, J., Liu, Y, Hu, B.X., Hao, Y, Huo, X., Fan, Y., Yeh, T. & Wang, Z.L. (2015). Analyzing effects of climate change on streamflow in a glacier mountain catchment using an ARMA model. Quaternary International, 358, 137-145.
14. Valipour, M. (2015). Long‐term runoff study using SARIMA and ARIMA models in the United States, Meteorological Applications, 22(3), 592-598.
15. Lohani, A. K., Kumar, R. & Singh, R.D. (2012). Hydrological time series modeling: A comparison between adaptive neuro-fuzzy, neural network and autoregressive techniques. Journal of Hydrology, 442, 23-35.
16. Unes, F., Demirci, M., Zelenakova, M., Calisici, M., Tasar, B., Vranay, F. & Kaya, Y. Z. (2020). River Flow Estimation Using Artificial Intelligence and Fuzzy Techniques. Water, 12(9), 2427.
17. Fathian, F., Mehdizadeh, S., Sales, A. K. & Safari, M. J. S. (2019). Hybrid models to improve the monthly river flow prediction: Integrating artificial intelligence and non-linear time series models. Journal of Hydrology, 575, 1200-1213.
18. Hussain, D. & Khan, A. A. (2020). Machine learning techniques for monthly river flow forecasting of Hunza River, Pakistan. Earth Science Informatics, 1-11.
19. Poonia, V. & Tiwari, H. L. (2020). Rainfall-runoff modeling for the Hoshangabad Basin of Narmada River using artificial neural network. Arabian Journal of Geosciences, 13(18), 1-10.
20. Partal, T. (2017). Wavelet regression and wavelet neural network models for forecasting monthly streamflow, Journal of Water and Climate Change, 8(1), 48-61.
21. Meyer, Y. (1993). Wavelets algorithms and applications. Society for Industrial and Applied Mathematics, Philadelphia.
22. Sahay, R. R. & Srivastava, A. (2014). Predicting monsoon floods in rivers embedding wavelet transform, genetic algorithm and neural network, Water Resources Management, 28(2), 301-317.
23. Mallat, S. G. (1989). A theory for multiresolution signal decomposition: the wavelet representation. IEEE transactions on pattern analysis and machine intelligence, 11(7), 674-693
24. Bayazıt, M. (1996). İnşaat mühendisliğinde olasılık yöntemleri. İstanbul Teknik Üniversitesi, İstanbul.
25. Hyndman, R. J. & Khandakar, Y. (2008). Automatic time series forecasting: the forecast package for R. Journal of Statistical Software, 26(3), 1–22.
26. Hyndman, R, Bergmeir, C., Caceres, G., Chhay, L., O'Hara-Wild M., Petropoulos, F., Razbash, S., Wang, E. & Yasmeen, F. (2018). Forecast: Forecasting functions for time series and linear models, R package version 8.3, http://pkg.robjhyndman.com/forecast (Erişimtarihi: 01.04.2018)
27. R Core Team, R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (Erişimtarihi: 01.04.2018).
28. Box, G. E. P. & Cox, D. R. (1964). An analysis of transformations. Journal of the Royal Statistical Society: Series B (Methodological), 26, 211–246.
29. Bickel, P. J. & Doksum K. A. (1981). An Analysis of Transformations Revisited. Journal of the American Statistical Association, 76, 296-311.
30. Wang, W & Ding J. (2003). Wavelet network model and its application to the prediction of hydrology, Nature and Science, 1(1), 67-71.