Analysis on Channel Cubage of the North Branch of the Yangtze River Estuary with BP Artificial Neural Network

Based on the hydrology and sediment data, an artificial neural network model which has been applied maturely and widely in all kinds of artificial neural network is established to study the relationship of channel cubage under 0m isobath in North Branch with the flow and sediment discharge at Datong gauging station, and the flow split ratio of the North Branch. The structure of the network model is fixed on 3-1-7-1. The network model is trained and tested well by choosing appropriate parameters; the computed results of BP artificial neural network agree well with that of multiple linear regressions. It can be concluded that BP artificial neural network may widely be used to predict the hydrological factors such as sediment discharge in river channel and estuary.