Computation of energy across the type-C piano key weir using gene expression programming and extreme gradient boosting (XGBoost) algorithm

An accurate assessment of energy loss across dams and weir systems is a critical technical and monetary remedy for understanding the hydraulic system's downstream morphology during the flood. Using the standard empirical formulas, accurately estimating the energy loss across the different hydraulic devices is a tedious and challenging procedure. Consequently, new and concise techniques remain highly sought after. This paper presents two empirical models based on the Gene Expression Programming (GEP) and Extreme Gradient Boosting (XGBoost) techniques to examine energy losses throughout the type-C PKW. The empirical models have been developed to consider five non-dimensional parameters, viz L/W, H-t /P, W-i/W-o, N, and S-i /S-o, that influence the energy over the weir significantly. The models were created using experimental data from a wide range of residual energy losses and release capacities. Additionally, the adequacy of the constructed GEP and XGBoost models was assessed using the RMSE (root mean square error) and statistical variables coefficient (R-2). As per the outcomes, the XGBoost model beats the GEP with the determination coefficient (R-2) = 0.999, MAE = 0.0062, MAPE = 1.4% and RMSE = 0.0012 in the training stage and R-2 = 0.998, MAE = 0.001, MAPE = 2.1, and RMSE = 0.001 in the testing data. These findings show that the XGBoost algorithm is more accurate than the two algorithms in this study for downstream energy prediction of type-C PKW. (c) 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under theCCBYlicense (http://creativecommons.org/licenses/by/4.0/).