Application of artificial neural networks in establishing regime channel relationships

The main purpose of this study is to evaluate the potential of simulating regime channel treatments using artificial neural networks. A collection of regime channel data with 371 data sets was collected from available literature. These data sets were randomly split into two subsets, i.e. Training and validation sets. The multi layer perceptron artificial neural network (MLP) was used to construct the simulation model based on the training data. The results show a considerably better performance of the NN model over the available empiric or rational equations. The constructed ANN models can almost perfectly simulate the width, depth and slope of alluvial regime channels. The values of correlation coefficient (R2) are close to one and the values of root mean square error (RMSE) are close to zero in all conditions. The results demonstrate that the ANN can precisely simulate the regime channel geometry, while the empirical, regression or rational equations can't.