Sustainable gene expression programming model for shear stress prediction in nonprismatic compound channels

Abstract

Open channel flow resistance and channel stability can be determined from boundary shear stress distribution. Overbank floods cross the main channel and pours the floodplain. The momentum change between the main channel and floodplains creates complicated flow shapes in compound channels. This alters floodplain and main channel shear stress. River floodplains also support agriculture and development activities due to which floodplain shape changes along the flow length, creating a converging, diverging or skewed compound channels. Traditional empirical formulas cannot accurately estimate shear force distribution. Thus, creative and accurate methods remain in demand. Gene expression programming, an innovative approach, has been utilized in this study to develop a new equation for compound channel with converging floodplains in terms of nondimensional geometric and flow variables to estimate the boundary shear force carried by floodplains. The proposed GEP-based method is best when compared to previously developed methods. The findings indicate that the predicted percentage shear force carried by floodplains determined using GEP is in good agreement with the experimental data (R2 = 0.96 and RMSE = 3.395 for the training data and R2 = 0.95 and RMSE = 4.022 for the testing data).