A nonlinear rainfall-runoff model embedded with an automated calibration method -Part 1 The model

Summary The purpose of this paper is to propose a nonlinear rainfall–runoff model that can overcome two problems of the ANN (artificial neural networks) based rainfall–runoff models. These two problems both result from the construction of the memory of the rainfall–runoff process. First, the two problems are explained. Then, the nonlinear computational unit (NCU), which is the building block of the proposed model, is introduced. The proposed model is established by cascading several NCUs and hence is called the nonlinear computational units cascaded (NCUC) model. The instructions to model the rainfall–runoff process using the NCUC model are given in this paper. Two actual storm events are simulated using a well-calibrated NCUC model. The modeling results are evaluated using five common performance criteria. The performance of the NCUC model is compared with that of three conventional and two ANN based rainfall–runoff models. The comparison shows that the NCUC model outperforms the conventional models and performs as well as the ANN based models. However, unlike the ANN based models, the memory of the rainfall–runoff process is involved implicitly in the NCUC model. Therefore, the capability of the NCUC model is more comprehensive than that of the ANN based models. It can be concluded that the NCUC model can solve the problems encountered by the ANN based rainfall–runoff models.

[1]  R. K. Kachroo.  River flow forecasting. Part 1. A discussion of the principles , 1992 .

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

[3]  Henrik Madsen,et al.  Comparison of different automated strategies for calibration of rainfall-runoff models , 2002 .

[4]  Jason Smith,et al.  Neural-Network Models of Rainfall-Runoff Process , 1995 .

[5]  E. Todini.  The ARNO rainfall-runoff model , 1996 .

[6]  S. Lallahem,et al.  A nonlinear rainfall-runoff model using neural network technique: example in fractured porous media , 2003 .

[7]  Vijay P. Singh,et al.  MIKE 11 - a generalized river modelling package. , 1995 .

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

[9]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[10]  V. Singh,et al.  Computer Models of Watershed Hydrology , 1995 .

[11]  U. C. Kothyari,et al.  Modeling of the daily rainfall-runoff relationship with artificial neural network , 2004 .

[12]  A. W. Minns,et al.  Artificial neural networks as rainfall-runoff models , 1996 .

[13]  Gwo-Fong Lin,et al.  A non-linear rainfall-runoff model using radial basis function network , 2004 .

[14]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[15]  B. S. Thandaveswara,et al.  A non-linear rainfall–runoff model using an artificial neural network , 1999 .

[16]  Li-Chiu Chang,et al.  Real‐time recurrent learning neural network for stream‐flow forecasting , 2002 .

[17]  Marco Franchini,et al.  Physical interpretation and sensitivity analysis of the TOPMODEL , 1996 .

[18]  Ashish Sharma,et al.  A study of optimal model lag and spatial inputs to artificial neural network for rainfall forecasting , 2000 .

[19]  Yen-Ming Chiang,et al.  A two‐step‐ahead recurrent neural network for stream‐flow forecasting , 2004 .