Application of the SVR-NSGAII to Hydrograph Routing in Open Channels

Abstract

Flow routing is used to simulate or predict downstream hydrographs on the basis of the features of upstream flow hydrographs. This paper combines support vector regression (SVR) and the nondominated sorting genetic algorithm II (NSGAII) into a hybrid hydrologic routing model called SVR-NSGAII in this paper for the prediction of a downstream flow hydrograph in simple and compound channels. The SVR-NSGAII hydrologic routing predictions are compared with those from hydraulic models in simple and compound channels. This paper’s results indicate that the SVR-NSGAII predicts the downstream hydrograph flow in a simple and compound channel, with approximately 94 and 98% accuracy, respectively.