مقایسه عملکرد ماشین بردار پشتیبان با سایر مدل‌های هوشمند در شبیه‌سازی فرآیند بارش- رواناب

     Simulation of rainfall-runoff process is a major step in water engineering studies and water resources management. In this study, the rainfall-runoff process of the Siminehroud monthly (1377-1390) were simulated using Support Vector Machines (SVM)  with Radial Basis kernel Function, Polynomial and linear Bayesian Network (BN) with a PC Learning Algorithm, also conventional methods such as Artificial Neural Networks (ANNs) and Gene Expression Programming (GEP) were used; finally, the results were compared with each other. Correlation Coefficient (CC), Root Mean Square Error (RMSE) and Nash-Sutcliff coefficient (NS) were used to evaluate the performance of the models. The results indicate the acceptable performance of the models and GEP model shows the highest CC (CC = 0.91), minimum RMSE (RMSE = 1.3 m3/s) and NS = 0.82 in verification stage.