Runoff Forecasting Method and System Based on Interannual and Monthly Variation Characteristics of Hydrological Variables

Abstract

The invention discloses a radial flow forecasting method and system based on hydrology variable yearly and monthly variation characteristics; the method comprises the following steps: obtaining history radial flow data of a forecasting section, and extracting forecasting month yearly radial flow, monthly rainfall before the forecasting month and monthly radial flow before the forecasting month from the history radial flow data; inputting the forecasting month yearly radial flow into a weighted moving average model, thus obtaining the weighted moving forecasting monthly radial flow; inputting monthly rainfall before the forecasting month and monthly radial flow before the forecasting month into a GRNN nerve network model, thus obtaining a nerve network forecasting monthly radial flow; using the least square method to weight and couple the weighted moving forecasting monthly radial flow and the nerve network forecasting monthly radial flow, thus obtaining the final combined forecasting monthly radial flow. The obtained final combined forecasting monthly radial flow considers rainfall and radial flow data utilization rate, and can improve the radial flow forecasting precision.

Description

technical field [0001] The invention belongs to the field of hydrological forecasting in hydrology, and more specifically relates to a runoff forecasting method and system based on interannual and monthly variation characteristics of hydrological variables. Background technique [0002] A common method for mid- and long-term runoff forecasting is to use models such as mathematical statistics or artificial neural networks to mine laws from past meteorological and hydrological information to predict future changes in runoff. In practical application, the correlation coefficient method is used to select predictors from the runoff, rainfall, temperature and other related factors in the first few months of the forecast month, and multiple regression, artificial neural network, fuzzy reasoning and other methods are used as driving models to mine runoff and predictors linear and nonlinear relationships. For the above data-driven models, the more useful information is mined, the cl...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above defects or improvement needs of the prior art, the present invention provides a runoff forecasting method and system based on the interannual and monthly variation characteristics of hydrological variables, thus solving the problems of the prior art with low availability of effective information and low forecasting accuracy technical issues

Images