An Empirical Research For Rainfall And Runoff Forecasting Based On Intelligent Combination Model

Due to the significant role of water resources in human production and life and the scarcity of available total water resources,effective hydrological forecasting can provide a reliable reference for water resources management and rational planning.Therefore,the accurate prediction of hydrological data is particularly important.This paper mainly makes two empirical analyzes,forecasting the precipitation of four rainfall stations of Songbai Hydrological station in Shennongjia Forest Region of Hubei province and the runoff at the Changmabao Hydrological station of the Shule river,Gansu province,respectively.In the first experiment,a novel hybrid model named SSA-DA-SSA is developed to forecast rainfall.SSA-DA-SVR model is established based on singular spectrum analysis(SSA),dragonfly algorithm(DA)and support vector regression(SVR)method.Firstly,SSA is used for extracting the trend components of the hydrological data.Then,DA is employed to optimize the penalty parameter c and parameter g of kernel function of SVR.Finally,SVR is selected as the main algorithm to model the relationship of input and output in the nonlinear system.The proposed SSA-DA-SVR method is used to forecast the monthly precipitation for Songbai,Panshui,Lanma and Jiulongchi stations.To validate the efficiency of the method,four compared models,DA-SVR,SSA-GWO-SVR,SSA-PSO-SVR,SSA-CS-SVR are established.The result shows the proposed method has the best performance among all five models,and its prediction has high precision and accuracy.The second experiment is mainly to forecast the monthly average runoff of Changmabao Hydrological station in the Shule river,Gansu Province.First,EEMD is used to decompose the original data in order to fully extract each trend.Then,according to the extracted trend features,each IMF is forecasted by the appropriate statistical models.Finally,the forecasting results of the IMFs are summed as the final result of the runoff,which is the predicted value of monthly average runoff by the combined model EEMD-IMFs-COM.In order to verify the efficiency of the proposed hybrid method,compared models are established.The result shows the proposed method has the best performance among compared models,and their prediction has high precision and accuracy.Through case analysis,the above two hybrid models can be used as effective and simple tools for hydrological forecasting.