Estimating Suspended Sediment Concentration by a Neaural Differential Evolution and Comparision it with ANFIS and RBF Models

In this study, neural differential evolution (NDE) models were used to estimate suspended sediment concentration. NDE models are improved by combining two methods, neural networks and differential evolution. At the first part of the study, the neural differential evolution is trained using daily river flow and suspended sediment data belonging to Givi Chay River at the northwest of Iran and various combinations of current daily stream flows, past daily stream flows and suspended sediment data are used as inputs to the neural differential evolution model so as to estimate current suspended sediment. In the second part of the study, the suspended sediment estimations provided by NDE model are compared with adaptive neuro- fuzzy inference system (ANFIS) and radial basis function (RBF) results. The Root mean squared error (RMSE) and the determination coefficient (R2) are used as comparison criteria. Obtained results demonstrate that NDE and ANFIS are in good agreement with the observed suspended sediment concentration; while they depict better results than RBF methods. For example, in Givi Chay River station, the determination coefficient (R2) is 0.9586 for NDE model, while it is 0.9152 and 0.8872 for ANFIS and RBF models, respectively. However, for the estimation of maximum sediment peak, the NDE was mostly found to be better than the ANFIS and the other techniques. The results also indicate that the NDE may provided better performance than the ANFIS and RBF in the estimation of the total sediment load (Re= -47%).