Investigating the extrapolation performance of neural network models in suspended sediment data

ABSTRACT Suspended sediment modelling is a quite significant issue in hydrology. The prediction of suspended sediment has taken the attention of several scientists in water resources. With extrapolation, the forecasting ability of the employed forecasting model beyond the calibration range is investigated. In the present study, different smoothing parameters are used to differentiate the kurtosis of the local critical points (local minima and maxima). The two models used are an artificial neural network (ANN) model and a multiple linear regression (MLR) model for prediction in order to examine the model extrapolation ability. The ANN model provides closer estimations to the observed peaks, being higher than the corresponding MLR ones. For the local minima, the ANN predictions are higher than the MLR predictions. As there are limited local points, all the remaining ANN predictions are lower than the MLR ones except for one point.

[1]  Jeffrey S. Simonoff,et al.  Handbook of Regression Analysis , 2012 .

[2]  Ali Osman Pektas,et al.  Creating a non-linear total sediment load formula using polynomial best subset regression model , 2016 .

[3]  Ozgur Kisi,et al.  Flow prediction by three back propagation techniques using k-fold partitioning of neural network training data , 2005 .

[4]  Mustafa Demirci,et al.  Prediction of suspended sediment in river using fuzzy logic and multilinear regression approaches , 2012, Neural Computing and Applications.

[5]  M. Sabouhi,et al.  Compartion of ANFIS, ANN, GARCH and ARIMA Techniques to Exchange Rate Forecasting , 2009 .

[6]  Jalal Chachi,et al.  Suspended load estimation using L1-fuzzy regression, L2-fuzzy regression and MARS-fuzzy regression models , 2016 .

[7]  A. Ahmadi,et al.  Daily suspended sediment load prediction using artificial neural networks and support vector machines , 2013 .

[8]  K. Loague.  Rainfall-Runoff Modelling , 2010 .

[9]  Ali Osman Pektas,et al.  ANN hybrid model versus ARIMA and ARIMAX models of runoff coefficient , 2013 .

[10]  Hikmet Kerem Cigizoglu,et al.  Generalized regression neural network in modelling river sediment yield , 2006, Adv. Eng. Softw..

[11]  Turgay Partal,et al.  Estimation and forecasting of daily suspended sediment data using wavelet–neural networks , 2008 .

[12]  Hikmet Kerem Cigizoglu,et al.  Estimation, forecasting and extrapolation of river flows by artificial neural networks , 2003 .

[13]  Ozgur Kisi,et al.  Methods to improve the neural network performance in suspended sediment estimation , 2006 .

[14]  S. Woznicki,et al.  Bayesian Regression and Neuro-Fuzzy Methods Reliability Assessment for Estimating Streamflow , 2016 .

[15]  Jeffrey G. Arnold,et al.  Evaluation of Three Watershed‐Scale Pesticide Environmental Transport and Fate Models 1 , 2007 .

[16]  M. Vafakhah,et al.  Evaluating the support vector machine for suspended sediment load forecasting based on gamma test , 2016, Arabian Journal of Geosciences.

[17]  Hikmet Kerem Cigizoglu,et al.  Suspended sediment load simulation by two artificial neural network methods using hydrometeorological data , 2007, Environ. Model. Softw..

[18]  H. K. Cigizoglu,et al.  Incorporation of ARMA models into flow forecasting by artificial neural networks , 2003 .

[19]  T. McMahon,et al.  Evaluation of automated techniques for base flow and recession analyses , 1990 .

[20]  Günter Blöschl,et al.  Spatio-temporal variability of event runoff coefficients , 2006 .

[21]  Hikmet Kerem Cigizoglu,et al.  Rainfall-Runoff Modelling Using Three Neural Network Methods , 2004, ICAISC.

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

[23]  H. K. Cigizoglu,et al.  ESTIMATION AND FORECASTING OF DAILY SUSPENDED SEDIMENT DATA BY MULTI-LAYER PERCEPTRONS , 2004 .