Performance evaluation of artificial ıntelligence and heuristic regression methods for rainfall-runoff modelling: An application in Aksu Stream

Year 2022, , 744 - 751, 18.07.2022

Hasan Törehan Babacan , Ömer Yüksek , Fatih Saka

https://doi.org/10.28948/ngumuh.1079616

Cited By: 1

Abstract

Rainfall-Runoff Models were developed by using the daily total precipitation (P) and daily mean discharge (Q) values on Aksu Stream and future discharges were predicted. The performances of numerical methods to explain the Rainfall-Runoff process were compared. Two kinds of Artificial Intelligence (AI) methods, including, Artificial Neural Networks (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS); as well as Heuristic Regression (HR) method, including Multivariate Adaptive Regression Splines (MARS) and Support Vector Machine (SVM), were used to describe the process. The performance of the methods was determined by using the Root Mean Square Error (RMSE), Correlation Coefficient (R) and Mean Absolute Error (MAE) coefficients. Although the performance of AI methods was very close, the lowest error value was obtained by ANFIS method. For the other methods, the highest performance was observed with MARS. It has been concluded that MARS method, which is HR method have higher performance than AI methods.

Keywords

Rainfall-Runoff Modelling, Artificial Intelligence, Heuristic Regression, Aksu Stream

Yapay zeka ve sezgisel regresyon yöntemlerinin yağış-akış modellemesi için performans değerlendirmesi: Aksu Deresi için bir uygulama

Abstract

Bu çalışmada, Aksu Deresi üzerinde, günlük toplam yağış (P) ve günlük ortalama debi (Q) değerleri kullanılarak Yağış-Akış Modelleri geliştirilmiş ve gelecek debi tahmini yapılmıştır. Yağış-Akış modellerinde kullanılan sayısal yöntemlerin Yağış-Akış sürecini açıklayabilme performansları kıyaslanmıştır. Yağış-Akış sürecini açıklamada, Yapay Zeka (AI) yöntemlerinden, Yapay Sinir Ağları (YSA) ve Adaptif Sinirsel Bulanık Çıkarım Sistemi (ANFIS) kullanılmıştır. Ayrıca Sezgisel Regresyon (HR) yöntemi olarak Çok Değişkenli Uyarlanabilir Regresyon Eğrileri (MARS) ve Destek Vektör Makineleri (SVM) kullanılmıştır. Yöntemlerin performansı Hata Ortalamalarının Karesinin Karekökü (RMSE), Korelasyon Katsayısı (R) ve Mutlak Hataların Ortalaması (MAE) katsayıları ile belirlenmiş ve kıyaslanmıştır. Denenen AI yöntemlerinin performansları çok yakın olsa da ANFIS yöntemi ile oluşturulan Yağış-Akış modelinde en düşük hata değeri elde edilmiştir. Denenen diğer yöntemlerde ise en yüksek performans değeri MARS ile oluşturulan Yağış-Akış modelinde gözlenmiştir. AI, HR ve diğer yöntemlerde yapılan performans değerlendirmesinde, MARS yönteminin AI yöntemlerine göre Yağış-Akış sürecini açıklayabilme performansının daha yüksek olduğu görülmüştür.

Keywords

Yağış-Akış Modellemesi, Yapay Zeka, Sezgisel Regresyon, Aksu Deresi

References

• D. Hughes, P. Greenwood, G. Coulson and G. Blair, Gridstix: Supporting flood prediction using embedded hardware and next generation grid middleware. International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM'06), pp. 6-12, New York, 2006.
• A. S. Tokar and P. A. Johnson, Rainfall-runoff modeling using artificial neural networks. Journal of Hydrologic Engineering, 4, 232-239, 1999. https://doi.org/10.1061/(ASCE)1084-0699(1999)4:3(2 32).
• G. Blöschl, M. F. Bierkens, A. Chambel, C. Cudennec, G. Destouni, A. Fiori and M. Renner. Twenty-three unsolved problems in hydrology (UPH)–a community perspective. Hydrological Sciences Journal, 64(10), 1141-1158, 2019. https://doi.org/10.1080/ 02626667.2019.1620507.
• P. K. T. Mishra and S. Karmakar, Performance of optimum neural network in rainfall–runoff modeling over a river basin. International Journal of Environmental Science and Technology, 16(3), 1289-1302, 2019. https://doi.org/10.1007/s13762-018-1726-7.
• M. Peel and T. A. McMahon, Historical development of rainfall‐runoff modeling. Wiley Interdisciplinary Reviews: Water, 7(5), 1471-1486, 2020. https://doi.org/10.1002/wat2.1471.
• J. Sitterson, C. Knightes, R. Parmar, K. Wolfe and M. Muche, An overview of rainfall-runoff model types. 9th International Congress on Environmental Modelling and Software, pp. 41-50, Colorado, USA, 24-28 Jun 2018.
• M. Ashrafi, L. H. Chua and C. Quek, The applicability of Generic Self-Evolving Takagi-Sugeno-Kang neuro-fuzzy model in modeling rainfall–runoff and river routing. Hydrology Research, 50(4), 991-1001, 2019.https://doi.org/10.2166/nh.2019.146.
• Ö. Yüksek, H. T. Babacan ve F. Saka, Yağış-akış modellemesinde optimum yapay sinir ağı yapısının araştırılması. Türk Hidrolik Dergisi, 2(1), 31-37, 2018.
• R. M. Adnan, A. Petroselli, S. Heddam, C. A. Santos and O. Kisi, Comparison of different methodologies for rainfall–runoff modeling: machine learning vs conceptual approach. Natural Hazards, 105(3), 2987-3011, 2021. https://doi.org/10.1007/s11069-020-04438-2.
• M. Jahandideh-Tehrani, G. Jenkins and F. A. Helfer, Comparison of particle swarm optimization and genetic algorithm for daily rainfall-runoff modelling: a case study for Southeast Queensland, Australia. Optimization and Engineering, 22(1), 29-50, 2021. https://doi.org/10.1007/s11081-020-09538-3.
• A. Talei, L. H. C. Chua and T. S., Wong, Evaluation of rainfall and discharge inputs used by Adaptive Network-based Fuzzy Inference Systems (ANFIS) in rainfall–runoff modeling. Journal of Hydrology, 391(3-4), 248-262, 2010. https://doi.org/10.1016/ j.jhydrol.2010.07.023.
• M. Kwon, H. H. Kwon and D. A. Han, hybrid approach combining conceptual hydrological models, support vector machines and remote sensing data for rainfall-runoff modeling. Remote Sensing, 12(11), 1801-1822, 2020. https://doi.org/10.3390/rs12111801.
• J. Adamowski, H. F. Chan, S. O. Prasher and V. N. Sharda, Comparison of multivariate adaptive regression splines with coupled wavelet transform artificial neural networks for runoff forecasting in Himalayan micro-watersheds with limited data. Journal of Hydroinformatics, 14(3), 731-744, 2012. https://doi.org/10.2166/hydro.2011.044.
• DSİ 2020 yılı resmi su kaynakları istatistikleri, https://www.dsi.gov.tr/Sayfa/Detay/1499 Erişim: 13.02.2022
• P. Hosseinzadeh Talaee, Multilayer perceptron with different training algorithms for streamflow forecasting. Neural Computing and Applications, 24(3), 695-703, 2014. https://doi.org/10.1007/s00521-012-1287-5.
• W. Wang, P. H. Van Gelder, J. Vrijling and J. Ma, Forecasting daily streamflow using hybrid ANN models. Journal of Hydrology, 324(1-4), 383-399, 2006. https://doi.org/10.1016/j.jhydrol.2005.09.032.
• H. T. Babacan, İklim değişikliğinin akarsu akışları üzerindeki etkisinin araştırılması: Doğu Karadeniz Havzası Örneği, Doktora Tezi, Karadeniz Teknik Üniversitesi Fen Bilimleri Enstitüsü, Türkiye, 2021.
• J. S. Jang, ANFIS: adaptive-network-based fuzzy inference system. IEEE transactions on systems, man, and cybernetics, 23(3), pp. 665-685, 1993.https://doi.org/10.1109/21.256541.
• N. N. Ozyurt and E. A. Sezer, Daily streamflow prediction by ANFIS modeling: Application to lower zamanti karst basin, Turkey. Journal of Intelligent and Fuzzy Systems, 23(6), 305-311, 2012.
• M. Kabir and M. M. J. Kabir, "Fuzzy membership function design: An adaptive neuro-fuzzy inference system (ANFIS) based approach, 2021 International Conference on Computer Communication and Informatics (ICCCI), pp. 1-5, Narashino, Japan, 27-29 Jan 2021. https://doi.org/10.1109/ICCCI50 826.2021.9402633.
• J. H. Friedman, Multivariate adaptive regression splines. The Annals of Statistics, 19(1), 1-67, 1991. https://doi.org/10.1214/aos/1176347963.
• C. Cortes and V. Vapnik, Support-vector networks, Machine Learning, 20(3), 273-297, 1995. https://doi.org/10.1007/BF00994018.
• H. Arslan, F. Üneş, M. Demirci, B. Taşar, ve A. Yılmaz, Keban baraj gölü seviye değişiminin ANFIS ve destek vektör makineleri ile tahmini. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 3(2), 71-77, 2020. https://doi.org/10.47495/okufbed.748018.
• S. K. Himanshu, A. Pandey and B. Yadav, Assessing the applicability of TMPA-3B42V7 precipitation dataset in wavelet-support vector machine approach for suspended sediment load prediction. Journal of Hydrology, 550, 103-117, 2017. https://doi.org/10.1016/j.jhydrol.2017.04.051.
• S. Ayhan ve Ş. Erdoğmuş, Destek vektör makineleriyle sınıflandırma problemlerinin çözümü için çekirdek fonksiyonu seçimi. Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 9(1), 175-201, 2014.
• M. A. Ghorbani, R. Khatibi, A. Goel, M. H. FazeliFard and A. Azani, Modeling river discharge time series using support vector machine and artificial neural networks. Environmental Earth Sciences, 75(8), 1-13, 2016. https://doi.org/10.1007/s12665-016-5435-6.
• A. Lal and B. Datta, Development and implementation of support vector machine regression surrogate models for predicting groundwater pumping-induced saltwater intrusion into coastal aquifers. Water Resources Management, 32(7), 2405-2419, 2018. https://doi.org/10.1007/s11269-018-1936-2.