Long-Term Runoff Modeling Using Rainfall Forecasts with Application to the Iguaçu River Basin

This work presents the development of a rainfall-runoff model for the Iguaçu River basin in southern Brazil. The model was developed to support the operation planning of hydroelectric power plants and is intended to predict the natural flow based on meteorological rain forecasts. A recurrent fuzzy system model was employed with parameters estimated by a genetic algorithm using observed rainfall as input. The model performs well using observed rainfall as input; however, its performance using predicted rainfall as input decays with the forecasting horizon, illustrating the effect of meteorological prediction errors. The prototype implementing the model has been used for dispatch planning by the Brazilian Electric System Operator.

[1]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[2]  C. Dahl,et al.  The Brazilian electrical system reform , 1999 .

[3]  Alberto Montanari,et al.  AFFDEF: A spatially distributed grid based rainfall-runoff model for continuous time simulations of river discharge , 2007, Environ. Model. Softw..

[4]  Jie Zhang,et al.  Recurrent neuro-fuzzy networks for nonlinear process modeling , 1999, IEEE Trans. Neural Networks.

[5]  M.E.P. Maceira,et al.  Periodic auto-regressive streamflow models applied to operation planning for the Brazilian hydroelectric system. , 2005 .

[6]  Paris A. Mastorocostas,et al.  A recurrent fuzzy-neural model for dynamic system identification , 2002, IEEE Trans. Syst. Man Cybern. Part B.

[7]  N. J. DE VOS,et al.  Multi-objective performance comparison of an artificial neural network and a conceptual rainfall—runoff model , 2007 .

[8]  Nelson F. F. Ebecken,et al.  Parameter Identification of Recurrent Fuzzy Systems With Fuzzy Finite-State Automata Representation , 2008, IEEE Transactions on Fuzzy Systems.

[9]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[10]  Yen-Ming Chiang,et al.  Comparison of static-feedforward and dynamic-feedback neural networks for rainfall -runoff modeling , 2004 .

[11]  E. Toth,et al.  Comparison of short-term rainfall prediction models for real-time flood forecasting , 2000 .

[12]  Robin T. Clarke,et al.  Medium-range reservoir inflow predictions based on quantitative precipitation forecasts , 2007 .

[13]  Yongkang Xue,et al.  Validation of the coupled Eta/SSiB model over South America , 2002 .

[14]  Ahmed El-Shafie,et al.  A neuro-fuzzy model for inflow forecasting of the Nile river at Aswan high dam , 2007 .

[15]  Secundino Soares,et al.  Optimal operation of reservoirs for electric generation , 1991 .

[16]  Sylviane Gentil,et al.  Recurrent neuro-fuzzy system for fault detection and isolation in nuclear reactors , 2005, Adv. Eng. Informatics.

[17]  Robert Babuska,et al.  Neuro-fuzzy methods for nonlinear system identification , 2003, Annu. Rev. Control..

[18]  K. P. Sudheer,et al.  A neuro-fuzzy computing technique for modeling hydrological time series , 2004 .

[19]  Soichi Nishiyama,et al.  Neural Networks for Real Time Catchment Flow Modeling and Prediction , 2007 .

[20]  Asaad Y. Shamseldin,et al.  A non-linear combination of the forecasts of rainfall-runoff models by the first-order Takagi–Sugeno fuzzy system , 2001 .

[21]  Meng Joo Er,et al.  NARMAX time series model prediction: feedforward and recurrent fuzzy neural network approaches , 2005, Fuzzy Sets Syst..

[22]  Nelson F. F. Ebecken,et al.  Identification of recurrent fuzzy systems with genetic algorithms , 2004, 2004 IEEE International Conference on Fuzzy Systems (IEEE Cat. No.04CH37542).

[23]  K.-Peter Holz,et al.  Short-term water level prediction using neural networks and neuro-fuzzy approach , 2003, Neurocomputing.

[24]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .