Prediction of water flow depth with kinematic wave equations and NARMAX approach based on neural networks in overland flow model

This paper deals with predict the water flow depth in presence of exceptional rain by two methods and to compare the performances of them based on error measurements. Two approaches are used and then compared: The knowledge modeling based on kinematic wave approach and the NARMAX (Nonlinear AutoRegressive Moving Average with eXogenous inputs) neural networks approach, based on real data taken from the Tondi Kiboro catchment area of Niger by a group from the IGE laboratory. For the first approach, the aim is to minimize the error between the measured and calculated flow rate values and to use the values of the parameters estimated during the optimization problem to calculate the new water flow depth values. For the second approach which is an unconventional method based on neural networks, the attempt is made to estimate the flow rate values using a recursive relation of the NARMAX approach through the use of a supervised learning model.

[1]  Miss A.O. Penney.  (b) , 1974, The New Yale Book of Quotations.

[2]  C. Micchelli.  Interpolation of scattered data: Distance matrices and conditionally positive definite functions , 1986 .

[3]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[4]  David S. Broomhead,et al.  Multivariable Functional Interpolation and Adaptive Networks , 1988, Complex Syst..

[5]  S. Galle,et al.  Overland flow and infiltration modelling for small plots during unsteady rain: numerical results versus observed values , 2000 .

[6]  Gail D. Baura,et al.  Nonlinear System Identification , 2002 .

[7]  Pierre Borne.  Les réseaux de neurones. , 2006 .

[8]  Abdu,et al.  Use of Horton Infiltration Model in Estimating Infiltration Characteristics of an Alfisol in the Northern Guinea Savanna of Nigeria , 2011 .

[9]  S. Billings,et al.  IDENTIFICATION OF NARMAX AND RELATED MODELS , 2011 .

[10]  Lin Ma,et al.  Prognostic modelling options for remaining useful life estimation by industry , 2011 .

[11]  A. Batailly,et al.  Interpolation et approximation de données à l'aide de courbes et surfaces paramétriques de type B-splines , 2012 .

[12]  Didier Georges,et al.  Optimal state estimation in an overland flow model using the adjoint method , 2014, 2014 IEEE Conference on Control Applications (CCA).

[13]  V. Nguyen.  Adjoint-based approach for estimation & sensor location on 1D hyperbolic systems with applications in hydrology & traffic , 2016 .

[14]  Chandranath Chatterjee,et al.  A wavelet-based non-linear autoregressive with exogenous inputs (WNARX) dynamic neural network model for real-time flood forecasting using satellite-based rainfall products , 2016 .

[15]  G. Besançon,et al.  Adjoint-based state and distributed parameter estimation in a switched hyperbolic overland flow model , 2016 .

[16]  Mustapha Ouladsine,et al.  Fault prognosis for batch production based on percentile measure and gamma process: Application to semiconductor manufacturing , 2016 .