Neural Network Model to Predict Characteristics of Hydraulic Jump in Stilling Basins with Convergent Wall

In this research، an Artificial Neural Network (ANN)، was adapted to model the hydraulic jump surface profile of the stilling basin with convergent wall. More than 1500 experimental data on depths of hydraulic jumps for basins of rectangular sections with 2. 7%، 4%، 5. 3% converged walls، were used. In developing ANN، 10 configurations، each having different number of hidden layers and/or neurons، were investigated. In each case، a configuration with attained the highest R2 value was selected as the optimal model. For stilling basin with convergent wall، the best ANN model for hydraulic jump profile was obtained with a 5-3-1 configuration، having 15 neuron in each of hidden layer and R2 =0. 999. High values of R2 obtained in all cases، suggesting a close agreement between the ANN،s output variable and the experimental data