End depth computation in inverted semicircular channels using ANNs

Abstract

The paper presents the application of artificial neural network (ANN) to determine the end-depth-ratio (EDR) for a smooth inverted semicircular channel in all flow regimes (subcritical and supercritical). The experimental data were used to train and validate the network. In subcritical flow, the end depth is related to the critical depth, and the value of EDR is found to be 0.705 for a critical depth–diameter ratio up to 0.40, which agrees closely with the value of 0.695 given by Dey [Flow Meas. Instrum. 12 (4) (2001) 253]. On the other hand, in supercritical flow, the empirical relationships for EDR and non-dimensional discharge with the non-dimensional streamwise slope of the channel are established.