Predicting the velocity field in a 90° Open channel bend using a gene expression programming model

In the present study, a gene expression programming (GEP) model is used to predict the velocity field in a 90° channel bend. By defining six different models, the effects of various input combinations on channel bend velocity prediction are investigated. An extensive experimental study was done on a 90° channel bend to evaluate the performance of the models. Experiments were conducted at 5 different discharges to obtain 325 depth-averaged velocity data. The comparison results indicate that the GEP with root mean square error of 0.0236, correlation coefficient of 0.9676 and average absolute deviation of 4.4685 performed well in velocity field prediction in bends. Additionally, an explicit equation to evaluate channel bend velocity was proposed, and it is applicable in practical cases.