Prediction of Sediment Concentration Using Artificial Neural Networks

Estimation of sediment concentration in rivers is very important for water resources projects planning and managements. In the literature, most of the sediment transport equations do not agree with each other and require many detailed data on the flow and sediment characteristics. The main purpose of the study is to establish an effective model which includes nonlinear relations between dependent (suspended sediment concentration) and independent (bed slope, flow discharge and sediment particle size) variables. Because of the complexity of the phenomena, a soft computing method artificial neural network (ANNs) which is the powerful tool for input-output mapping is used for estimating total sediment load concentration. fit the present study. 60 experiments were used for establishing ANN model. However, ANN model was compared with some sediment transport equations. The results show that ANN model is found to be significantly superior to others. The ANN model performs best followed by the model of Modified Einstein Formula (Einstein-Brown) and also results of Modified Einstein Formula are in agreement with observed data and ANN model. The results of Graf and Acaroglu Formulae however, were not found to be ill agreement with the observed data.