Forecasting suspended sediment load using regularized neural network: Case study of the Isser River (Algeria)

In the management of water resources in different hydro-systems it is important to evaluate and predict the sediment load in rivers. It is difficult to obtain an effective and fast estimation of sediment load by artificial neural network without avoiding over-fitting of the training data. The present paper comprises the comparison of a multi-layer perception network once with non-regularized network and the other with regularized network using the Early Stopping technique to estimate and forecast suspended sediment load in the Isser River, upstream of Beni Amran reservoir, northern Algeria. The study was carried out on daily sediment discharge and water discharge data of 30 years (1971–2001). The results of the Back Propagation based models were evaluated in terms of the coefficient of determination (R2) and the root mean square error (RMSE). Results of the comparison indicate that the regularizing ANN using the Early Stopping technique to avoid over-fitting performs better than non-regularized networks, and show that the overtraining in the back propagation occurs because of the complexity of the data introduced to the network