Application of Artificial Neural Network to Estimate Hydraulic Jump Characteristics in Divergent Rectangular Sections on Inverse Slope

Stilling basins are the most important energy dissipating structures placed at the end of spillways and any source of supercritical flow to control the hydraulic jump. Due to its importance and high construction costs، modeling of stilling basins are necessary prior to construction. Physical modeling of stilling basins are time consuming and costly، therefore attempts have been made so far to relate the hydraulic jump characteristics such as the jump length، energy loss، etc.، to some hydraulic parameters like Froude number، divergence and the bed slope. In this study hydraulic jump characteristics such as the jump length and energy loss in divergent rectangular sections on inverse slopes were modeled as functions of Froude number، angle of divergence and inverse bed slope، using Artificial Neural Network. The accuracy of the model for estimating different hydraulic parameters was also verified. The results indicated that the model was capable of predicting hydraulic parameters with high accuracy. Furthermore، the weight of each parameter for estimating hydraulic characteristics was determined. Data Fit software was used to produce relationships between the parameters. The relationships found to be accurate enough to predict the hydraulic jump characteristics.