Comparative Analysis of Rainfall-Runoff Modeling Using Support Vector Machines for Two Dams in Uttarakhand

Abstract

The main objective of this study was to evaluate and compare the performances of rainfall-runoff models that were developed by using support vector machines (SVMs). Rainfall and runoff data of Haripura and Baur dams were adopted on daily basis from Irrigation Division Rudrapur in Uttarakhand. In this study, radial kernel function was used. As the values of Cost function (C), and varies, performances of the models can be altered. So, at optimum values of these variables, there exists a best correlation between rainfall and runoff. It can be inferred from the study that SVM models provide satisfactory results for both dams. These results can be used for runoff prediction for various purpose such as irrigation etc.