Estimation of Long- Shore Sediment Transport Using neural Networks

Estimation of long-shore sediment transport rate is one of the most common and difficult problems in coastal engineering. Ue to the importance of sediment behavior, ample atudy have been carried out in order to final a better correlation between the parameters. On the other hand, due to many variables involed, the equations offered by field and laboratory investigators are in a very wide range and sometimes quite different from each other. in this paper the use of neural networks for estimation of long-shore sediment transport rate is presented. In order to heve a good moodel for predicting sedimentation in coastal situations, the algoritm pf program must be trained with a lot of data. For this reason the field data expressed by Kamphuis (1986) has been used. Finally the outputs of neural network program have been compared with results of Kamphuis and other investigators. Good correlation betweer. ssults of neural network program and field data shows thet this method is better than all earliest relations and methodes.