Author(s): Breac Macleod; Alan Ervine
Linked Author(s):
Keywords: No Keywords
Abstract: This paper describes the development of an Artificial Neural Network (ANN) model to predict the conveyance or discharge capacity of meandering compound channels. An ANN model was trained to approximate the functional relationship mbetween discharge capacity and nine major parameters which influence flow behaviour in meandering compound channels. The training data which exhibits this functional relationship was collated from a range of experimental data sets which have been compiled world-wide. The final training set incoporates data from both large and small scale model experiments as well as field data. The resulting ANN model is at least as good, if not better than any deterministic models developed for this purpose.
Year: 1997