Daily Suspended Sediment Prediction Using Seasonal Time Series and Artificial Intelligence Techniques

Estimating the amount of suspended sediment in rivers correctly is important due to adverse impacts encountered during design and maintenance hydraulic structures such as dams, regulators, water channels bridges. The concentration discharge currents have usually complex relationship, especially on long term scales, which can lead high uncertainties load estimates for certain components. In this paper, with several data-driven methods, including two types perceptron support vector machines radial basis function kernel (SVM-RBF), poly learning algorithms (SVM-PK), Library SVM (LibSVM), adaptive neuro-fuzzy (NF) statistical approaches rating curves (SRC), multi linear regression (MLR) are used forecasting daily from temperature streamflow river. Daily data measured at Augusta station by US Geological Survey. 15 different input combinations (1 15) were SVM-PK, SVM-RBF, LibSVM, NF MLR model studies. All compared each other according three criteria; mean absolute errors (MAE), root square (RMSE) correlation coefficient (R). Of applied nonlinear LibSVM good results, but generates a slightly better fit under whole values.