Author(s): H. M. Nagy; K. Watanabe; M. Hirano
Linked Author(s):
Keywords: Sediment transport; Natural rivers; Neural networks; Back propagation algorithm
Abstract: Artificial neural networks is an advanced topic which provides hydraulic and environmental engineers with a strong tool for estimating missing information to be used for design purposes and management practice. In this study, a neural networks is used to estimate the natural sediment discharge in rivers in terms of sediment concentration. This is achieved by training the network to extrapolate data collected from reliable sources. Selecting an appropriate neural networks structure and a training algorithm as well as providing data to the network are addressed using a constructive algorithm called back-propagation algorithm (BPA). Sensitivity analysis is performed for flow and sediment parameters. The predicted sediment concentrations are agreed well with the measured ones.
Year: 1999