Simulation and prediction of scour whole dimensions downstream of siphon overflow using support vector machine and Gene expression programming algorithms

Background and Objectives: The purpose of this study is to simulate and predict the dimensions of the scour cavity downstream of the siphon overflow using the SVM model and compare it with other numerical methods. The use of the SVM algorithm as a meta-heuristic system in simulating complex processes in which the dependent variable is a function of several independent variables has been widely developed among researchers. The purpose of this study is to simulate and predict the dimensions of the scour cavity downstream of the siphon overflow using the SVM model and compare it with other numerical methods. The use of the SVM algorithm as a meta-heuristic system in simulating complex processes in which the dependent variable is a function of several independent variables has been widely developed among researchers. Methods: The innovative aspect of this research is the prediction and numerical comparison of the geometric dimensions of the scour cavity downstream of the siphon overflow in the form of using the SVM based data model with other numerical models. The measured values ​​were collected through the collection of laboratory data performed in the laboratory. Findings: Finally, using statistical indicators, the accuracy of each model is calculated and their function is classified. Conclusion: Performance evaluation indices R, RMSE and ZDDR (max) in test and testing processes of two intelligent SVM and GEP algorithms in simulating scour hole dimensions downstream of siphon overflow for three 30, 45 and 60 degree projectiles with materials Sediment with average particle sizes of 1.4, 3.7 and 8.1 mm indicates better performance of the GEP model than the SVM model.