River flow time series prediction with a range-dependent neural network

Artificial neural networks provide a promising alternative to hydrological time series modelling. However, there are still many fundamental problems requiring further analyses, such as structure identification, parameter estimation, generalization, performance improvement, etc. Based on a proposed clustering algorithm for the training pairs, a new neural network, namely the rangedependent neural network (RDNN) has been developed for better accuracy in hydrological time series prediction. The applicability and potentials of the RDNN in daily streamflow and annual reservoir inflow prediction are examined using data from two watersheds in China. Empirical comparisons of the predictive accuracy, in terms of the model efficiency R and absolute relative errors (ARE), between the RDNN, back-propagation (BP) networks and the threshold auto-regressive (TAR) model are made. The case studies demonstrated that the RDNN network performed significantly better than the BP network, especially for reproducing low-flow events.

[1]  Witold F. Krajewski,et al.  Rainfall forecasting in space and time using a neural network , 1992 .

[2]  Christian W. Dawson,et al.  An artificial neural network approach to rainfall-runoff modelling , 1998 .

[3]  Howell Tong,et al.  Threshold autoregression, limit cycles and cyclical data- with discussion , 1980 .

[4]  H. Raman,et al.  Multivariate modelling of water resources time series using artificial neural networks , 1995 .

[5]  G. Kember,et al.  Forecasting river flow using nonlinear dynamics , 1993 .

[6]  K. Thirumalaiah,et al.  River Stage Forecasting Using Artificial Neural Networks , 1998 .

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

[8]  R. E. Griffith,et al.  A Nonlinear Programming Technique for the Optimization of Continuous Processing Systems , 1961 .

[9]  Shie-Yui Liong,et al.  River Stage Forecasting in Bangladesh: Neural Network Approach , 2000 .

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

[11]  Bernard Bobée,et al.  Daily reservoir inflow forecasting using artificial neural networks with stopped training approach , 2000 .

[12]  Slobodan P. Simonovic,et al.  Short term streamflow forecasting using artificial neural networks , 1998 .

[13]  Kadir Liano,et al.  Robust error measure for supervised neural network learning with outliers , 1996, IEEE Trans. Neural Networks.

[14]  H. Tong.  On a threshold model , 1978 .

[15]  Stan Openshaw,et al.  A hybrid multi-model approach to river level forecasting , 2000 .

[16]  D. K. Srivastava,et al.  Application of ANN for Reservoir Inflow Prediction and Operation , 1999 .

[17]  Sevket Durucan,et al.  River flow prediction using artificial neural networks: generalisation beyond the calibration range. , 2000 .

[18]  A. Soldati,et al.  River flood forecasting with a neural network model , 1999 .

[19]  A. Shamseldin.  Application of a neural network technique to rainfall-runoff modelling , 1997 .

[20]  N. Null.  Artificial Neural Networks in Hydrology. I: Preliminary Concepts , 2000 .

[21]  Adriaan van den Bos,et al.  Nonlinear least-absolute-values and minimax model fitting , 1985, Autom..

[22]  T. E. Baker,et al.  Successive Linear Programming at Exxon , 1985 .

[23]  Geoffrey E. Hinton,et al.  Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[24]  J. David Fuller,et al.  Back propagation in time‐series forecasting , 1995 .

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

[26]  Arjen van Ooyen,et al.  Improving the convergence of the back-propagation algorithm , 1992, Neural Networks.

[27]  Scott E. Fahlman,et al.  An empirical study of learning speed in back-propagation networks , 1988 .

[28]  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..

[29]  W. E. Hart,et al.  A critique of current methods in hydrologic systems investigation , 1964 .

[30]  Manuel J. Rodriguez,et al.  Neural network-based modelling of the adequate chlorine dosage for drinking water disinfection , 1996 .

[31]  L. Natale,et al.  Non-linear modelling of the rainfall-runoff transformation , 1992 .

[32]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[33]  Linda See,et al.  Applying soft computing approaches to river level forecasting , 1999 .

[34]  Richard J. Heggen,et al.  Neural Networks for River Flow Prediction , 1995 .

[35]  Rao S. Govindaraju,et al.  Prediction of watershed runoff using Bayesian concepts and modular neural networks , 2000 .

[36]  Non-linear runoff routing — A comparison of solution methods , 1986 .

[37]  Konstantine P. Georgakakos,et al.  Chaos in rainfall , 1989 .

[38]  H. Tong.  Non-linear time series. A dynamical system approach , 1990 .

[39]  Mark S. Seyfried,et al.  Searching for chaotic dynamics in snowmelt runoff , 1991 .

[40]  V. Chandramouli,et al.  Deriving a General Operating Policy for Reservoirs Using Neural Network , 1996 .

[41]  M. Lorrai,et al.  Neural nets for modelling rainfall-runoff transformations , 1995 .

[42]  T. V. Hromadka,et al.  Approximate confidence intervals for design floods for a single site using a neural network , 1999 .

[43]  David E. Rumelhart,et al.  Predicting the Future: a Connectionist Approach , 1990, Int. J. Neural Syst..

[44]  Amir F. Atiya,et al.  A comparison between neural-network forecasting techniques-case study: river flow forecasting , 1999, IEEE Trans. Neural Networks.

[45]  Marie Cottrell,et al.  Neural modeling for time series: A statistical stepwise method for weight elimination , 1995, IEEE Trans. Neural Networks.