Development Of Rainfall-Runoff Forecast Model
Keywords:
Feed forward network, Levenber Marquadt, Bayesian RegularisationAbstract
In recent time, Artificial Neural Network (ANN) has been found useful in solving engineering problems; its accuracy in forecast of rainfall-runoff for tropical region was investigated in this work. Development of three-layered feed-forward model for rainfall-runoff forecast using gauge height, rainfall and evaporation rates was considered. Levenberg Marquadt and Bayesian Regularisation were used in training the models with data sets from two selected hydrological gauging stations of BeninOwena River Basin Development Authority. Multiple Linear Regression model was also developed in order to compare its forecast accuracy with three-layered feedforward model. The results obtained from the models were evaluated using coefficient of determination and root mean square error as performance statistics. From the results, the model showed higher coefficient of determination and lower root mean square error for the three-layered feed-forward networks. It was concluded that the three-layered feed-forward model improved the forecast accuracy of the runoff of Benin-Owena river basin than multiple linear regression model using the same hydrological condition.
References
ASCE (2005). Artificial Neural Networks in hydrology-I: Preliminary Concepts Journal of Hydrological Engineering, ASCE Task Committee on Application of ANNS. Hydrology, 124-137.
Atiya A. F., El-Shoura S., Shaheen S. I. and El-Sherif M. (1999). A Comparison between Neural Network Forecasting Techniques: River flow Forecasting, IEEE Trans. Neural Networks, 10, 402- 409.
Aqil M., Ichiro Kita, Akira Yano and Soichi Nishiyama (2007). A comparative study of artificial neural networks and neuro-fuzzy in continuous modeling of the daily and hourly behaviour of runoff. Journal of hydrology, 337, 22-34.
Brath A. and Rosso R. (1993). Adaptive Calibration of a Conceptual Model for Flash Flood Forecasting. Water Resources Research, 29(8), 2561-2572.
Dekker L. W. and Ritsema C. J. (2003) Wetting patterns in water repellent Dutch soils. In Ritsema C.J.and Dekker L.W. (Eds.) Soil Water Repellency. Elsevier: Amsterdam, pp. 151-166.
Hsu K. Gupta, H.V. and Sorooshian, S. (1995). Artificial neural network modeling of the rainfallrunoff process. Water Resources Research, 31, 2517-2530.
Jang J. S. R., Sun C. T. and Mizutani E. (1997). Neuro-fuzzy and Soft Computing: A Computational Approach to Learning and Machine Intelligence. New Jersey: Prentice-Hall.
Markus M. and Baker D. (1994). The Fraser River: Streamflow forecasting and Simulation computer Package. Technical Report, Northern Colorado Water Conservancy District, Loveland.
McCuen R. H. and Snyder M. W. (1986). W. Hydrologic modelling: Statistical methods and applications. Englewood Cliffs, NJ: Prentice-Hall.
Morales H. K., Ibrahim G. J., Chen J. C. and Ryan M. L. (2006). Bayesian Model Averaging with applications to Benchmark Dose Estimation for Arsenic in Drinking Water. Journal of the American Statistical Association, 101, 473.
NayakP. C., Sudheer K. P. and Jain S. K. (2007). Rainfall-Runoff Modeling through Hybrid Intelligent System. Water Resources Research, 43, w07415, doi:10.1029/2006WR004930.
Smith, T. J. and Marshall, L. A. (2009). A Conceptual Precipitation-Runoff Modeling suite: Model Selection, Calibration and Predictive uncertainty Assessment. 18th World IMACS/MODSIM Congress, Cairns, Australia. alia
