Editorial Type:
Article Category: Research Article
 | 
Online Publication Date: 24 Feb 2011

Rainfall-Runoff Simulation Using Fuzzy Linear Regression—Case Study: Kasilian, a RepresentativeWatershed in Iran

and
Page Range: 212 – 226
DOI: 10.5555/arwg.10.3-4.qw565500155680x5
Save
Download PDF

This study focuses on evaluating fuzzy linear regression (FLR), the hydrological conceptual model (Nash model), and a black-box model (artificial neural network, or ANN) in order to simulate the rainfall-runoff process. Three different scenarios that include runoff and precipitation in previous time steps as independent variable and the corresponding runoff in a future time step as the dependent variable are considered, in which the regression coefficient includes fuzzy numbers with an asymmetric triangular membership function. Eight rainfall-runoff events in the Kasilian representative watershed in northern Iran were used to evaluating the performance of the above models. Finally, the FLR scenario that best simulates hydrographs is compared with the results of the Nash and ANN models. Results indicate that the FLR simulates the rising and falling limbs and overall shape of the hydrograph better than the other models, while the ANN model is more efficient in simulating the peak flow of the hydrograph.

Cet article évalue trois modèles pour simuler le processus de précipitation-écoulement dans un bassin versant : une régression linéaire floue, un modèle hydrologique (le modèle conceptuel de Nash), et un modèle de boîte noire (un réseau neural artificiel). Les trois scénarios examinés prennent l'écoulement et les précipitations dans une période donnée comme variables indépendantes et l'écoulement correspondant dans la période suivante comme variable dépendante. Le coefficient de régression intègre des nombres flous avec une correspondance triangulaire asymétrique. Huit épisodes de précipitations dans le bassin versant kasilien, dans le Nord de l'Iran, ont été utilisés pour tester l'efficacité de ces modèles. Le scénario employant la régression linéaire floue, qui simule les résultats correspondants le mieux aux hydrogrammes, a été comparé aux deux autres modèles. On note que la régression linéaire floue reprend les parties ascendantes et descendantes ainsi que la forme générale des hydrogrammes de manière plus satisfaisante que les autres modèles, tandis que le modèle de la boîte noire simule plus efficacement les écoulements maximaux.

ASCE Task Committee. 2000 Artificial neural networks in hydrology, 2: Hydrologic applications. ASCE Journal of Hydrologic Engineering 5:124-37.

Bardossy, A., and Duckstein, L. 1995. Fuzzy rule-based modeling with applications to geophysical, biological and engineering systems. Boca Raton, Florida: CRC Press.

Daka, P., and Chandramouli,V. 2005. Fuzzy neural network model for hydrologic flow routing. Journal of Hydrologic Engineering 10:302-14.

Fausett, L. 1994. Fundamentals of networks. Englewood Cliffs, New Jersey: Prentice Hall.

Flood, I., and Kartam, N. 1994. Neural networks in civil engineering, I: Principles and understanding. Journal of Computing in Civil Engineering 8:131-48.

Gupta, H.V., Hsu, K., and Sorooshian, S. 2000. Effective and efficient modeling for streamflow forecasting. In Artificial Neural Networks in Hydrology, ed. R. S. Govindaraju and A. R. Rao, 7-22. Dordrecht: Kluwer.

Hong D. H., Lee S., and Do H.Y. 2001. Fuzzy linear regression analysis for fuzzy inputoutput data using shape-preserving operations. Fuzzy Sets and Systems 122:513-26.

Kao C. and Chyu C. L. 2002.A fuzzy linear regression model with better explanatory power. Fuzzy Sets and Systems 126:401-9.

Kaufman, A., and Gupta, M. M. 1991. Introduction to fuzzy arithmetic: Theory and applications. New York:Van Nostrand Reinhold.

Lee, K. T., and Chang, C.-H. 2005. Incorporating subsurface-flow mechanism into geomorphology-based IUH modeling. Journal of Hydrology 311:91-105.

Nash, J. E. 1957. The form of instantaneous unit hydrograph. International Association Hydrology Science Publication (LAHS, Gentbrugge, Belgium) 51:546-57.

Ozelkan, E. C. 1997. Multi-objective fuzzy regression applied to the calibration of conceptual rainfall-runoff models. PhD dissertation, University of Arizona, Tucson, Arizona.

Ozelkan, E. C., and Duckstein, L. 2001. Fuzzy conceptual rainfall-runoff models. Journal of Hydrology 253:41-68.

Singh, V. P. 1988. Hydrolgic systems: Rainfall runoff modeling, volume 1. New Jersey: Prentice Hall.

Tanaka, H., Uejima S., and Asai K. 1982. Linear regression analysis with fuzzy model. IEEE Transaction Systems, Man and Cybernet 12:903-07.

Wu, J. S. P. E., Han, J., Annambhotla, S., and Bryant, S. 2005. Artificial neural networks for forecasting watershed runoff and stream flows. Journal of Hydrologic Engineering 10:216-22.

Yen K. K., Goshray, S., and Roig, G. 1999.A linear regression model using triangular fuzzy number coefficients. Fuzzy Sets and Systems 106:166-77.

  • Download PDF