A new hybrid forecasting approach applied to hydrological data: A case study on precipitation in northwestern China

Abstract

Hydrogeological disasters occur frequently. Proposing an effective prediction method for hydrology data can play a guiding role in disaster prevention; however, due to the complexity and instability of hydrological data, this is difficult. This paper proposes a new hybrid forecasting model based on ensemble empirical mode decomposition (EEMD), radial basis function neural networks (RBFN), and support vector machine (SVM), this is the EEMD-RBFN-SVM method, which has achieved effective results in forecasting hydrologic data. The data were collected from the Yushu Tibetan Autonomous Region of the Qinghai Province. To validate the method, the proposed hybrid model was compared to the RBFN, EEMD-RBFN, and SAM-ESM-RBFN models, and the results show that the proposed hybrid model had a better generalization ability.