Study On The Model Of Rainfall-runoff Forecasting Based On Improved L-M BP Network

The result of rainfall-runoff forecasting is required for effective hydropower reservoir management and scheduling. It can be widely used in many fields such as flood control, drought protecting, environment protecting, operation of reservoir, running of hydropower station, ship management and water resource distribution. It can show huge economic value in operation of reservoir hydropower station. But the precision of the result is not satisfactory to us because of the complexity of intrinsic flow mechanism and the influence of human activity to the river basin. So, new theory and methods are being urgently needed for the study on the rainfall-runoff forecasting. Under these conditions, the model based on the improved Levenberg-Marquardt BP network is applied in rainfall-runoff forecasting, and provides an effective approach to improve accuracy of rainfall-runoff forecasting. The focus of the paper combined with the project of SongHua River drainage area is as follows:(1) In order to improve Levenberg-Marquardt BP rainfall-runoff forecasting model, this paper uses the two-axis method to calculate average precipitation rain fall; and separates the flood season and low water season.(2) An improved Levenberg-Marquardt BP network model was developed to forecast river flow in the SongHua River drainage area. The result of the simulation and comparative experiment indicates that the network model proposed by this paper can be applied more successfully in the complicated river network.(3) A credible and applied hydraulic dynamics model based on Levenberg-Marquardt BP network is applied to SongHua River drainage area, and the model proposed by this paper can be applied higher accuracy and more convenience.