Application of Two Methods of Artificial Neural Network MLP, RBF for Estimation of Wind of Sediments (Case Study: Korsya of Darab Plain)

The lack of sediment gauging stations in the process of wind erosion, caused of estimate of sediment be process of necessary and important. Artificial neural networks can be used as an efficient and effective of tool to estimate and simulate sediments. In this paper two model multi-layer perceptron neural networks and radial neural network was used to estimate the amount of sediment in Korsya of Darab city. The simulations were performed with two methods common artificial neural network Initially, the sediment was sampled with sediment traps afterward  by climatic factors such as average wind speed, evaporation, precipitation, relative humidity, minimum temperature, maximum temperature, average temperature and Percentage of vegetation, respectively, as the dependent variable and independent input to the model was chosen to simulate the sediment. The Results of the performance model show that MLP neural network (feedforward Back propagation algorithm) with learning technique of Calibrated dual gradient Compared  to  radial  Neural Network Respectively with coefficient of determination .95 , Roots mean square error .02 and  coefficient of determination .90 , Roots mean square error .40 for  estimate of wind deposits  were the higher  efficiency and accurate. Course, it should be noted that an important advantage of the artificial neural network Multilayer Perceptron to Neural Network Radial is more flexibility.