Estimating scour below inverted siphon structures using stochastic and soft computing approaches

In this work, we use the non-linear regression, artificial neural network (ANN), and genetic programming (GP) approaches in order to predict an important tangible issue, i.e. the scour dimension downstream of inverted siphon structures. Dimensional analysis and non-linear regression-based equations are proposed for the estimation of the maximum scour depth, location of the scour hole, and location and height of the dune downstream of the structures. In addition, The GP-based formulation results are compared with the experimental results and other accurate equations. The analysis results show that the equations derived from the forward stepwise non-linear regression method have the correlation coefficients R2= 0.962, 0.971, and 0.991, respectively. This correlates the relative parameter of the maximum scour depth (s/z) in comparison with the GP and ANN models. Furthermore, the slope of the fitted line extracted from computations and observations for dimensionless parameters generally presents a new achievement for sediment engineering and scientific community, indicating the superiority of the ANN model.