An analysis of shear stress distribution in circular channels with sediment deposition based on Gene Expression Programming

Knowledge of the boundary shear stress distribution in channels is important because it is a key factor affecting on erosion and sedimentation rates. The presence of sediment deposits in sewers is often reported during operation, and circular channels are frequently used in sewer networks. Gene expression programming (GEP) is applied in this study to determine an equation for evaluating the shear stress distribution along the wetted perimeter of a circular channel with a flat bed, because of the presence of sediment on the bed. In view of the parameters affecting the shear stress distribution, five dimensionless parameters are applied to develop six GEP models to be used with 905 experimental data. The impact of the shear stress parameters is studied using the six GEP models and by dividing the wetted perimeter into wall and bed sections. Two equations are extracted from the GEP models' output to estimate wall and bed shear stresses. The best model results are compared with a well-known equation based on the entropy concept. The GEP model predictions of wall and bed shear stresses are very similar to the experimental outcomes, whereas the entropy-based model overestimates the shear stress distribution. The proposed GEP models demonstrate superior performance in estimating the shear stress distribution with a mean absolute percentage error (MAPE) of 3.79% compared to an existing equation with MAPE of 9.52%. (C) 2017 International Research and Training Centre on Erosion and Sedimentation/the World Association for Sedimentation and Erosion Research. Published by Elsevier B.V. All rights reserved.