Drops and screens are the commonest energy dissipator structures in irrigation networks and erodible canals. This research investigates the efficiency of intelligence methods in predicting relative energy dissipation on an inclined drop with a screen. To this end, in addition to the conventional support vector machine (SVM) model, a new advanced productivity prediction model of the SVM model coupled with the Harris hawks optimizer algorithm (SVM-HHO) was developed. In this study, 138 tests were conducted to investigate the relative energy dissipation with variable discharge, two drop heights, three different angles of a drop, and two porosity ratios of the screen in the laboratory. The performances of the proposed models were evaluated using statistical analysis containing coefficient of determination (R2), root-mean-square error (RMSE), Kling-Gupta efficiency (KGE), probability density function plot, scatter plot, estimation errors plot, and box and whisker plot. The results indicated that in addition to the successful performance of the inclined drop with a screen, the hybrid SVM-HHO method with R2 = 0.992, RMSE = 0.399, and KGE = 0.997 performs more precisely than the standalone SVM model with R2 = 0.977, RMSE = 1.435, and KGE = 0.893 in estimating relative energy dissipation on the inclined drop.
