Four-dimensional coupling medium and long-term runoff forecasting method for gas, land, reservoir and water

Abstract

The invention provides an air-land-reservoir-water four-dimensional coupled medium and long-term runoff forecasting method. A drainage basin air-land-reservoir-water calculation scheme based on a climate mode, a series-connection two-parameter water balance model, a storage capacity difference model and an incoming water synthesis model is established; a gas-land coupler, a land-reservoir coupler and a reservoir water coupler are established respectively, and through intelligent correction, a state judgment machine, uniform distribution and other methods, climate mode grid point output is converted into partition input of two parameter models, flow output of the two parameter models is converted into water level input of a reservoir model, and water storage output of the reservoir model is converted into influence flow input of an incoming water model; and coupling the models by utilizing a coupler, and establishing a medium-and-long-term runoff coupling forecasting model to obtain an actual flow forecast value. By means of the method, ten-day-month-season scale runoff forecasting can be achieved, the influence of reservoir regulation and storage on runoff forecasting is considered, the method is high in precision compared with a common method, and water balance can be guaranteed.

Description

technical field [0001] The invention relates to the technical field of medium and long-term runoff forecasting in river basins, in particular to a medium- and long-term runoff forecasting method with four-dimensional coupling of air and land reservoir water. Background technique [0002] Medium- and long-term runoff forecasting is an important technical means to achieve rational allocation and efficient utilization of water resources. Traditional medium- and long-term hydrological forecasting methods are generally based on the continuous, periodic, regional and random characteristics of river runoff changes. Method research, because the hydrological process is a complex and highly nonlinear process, some new modern artificial intelligence methods that can describe nonlinear information have been introduced into the field of long-term hydrological forecasting, such as deep learning, big data mining analysis, etc. However, the above methods mainly use climatic factors, histori...

AI Technical Summary

Problems solved by technology

However, the above methods mainly use climate factors, historical runoff, etc. The contradictory problem of more rainfall forecast but less flow forecast

At the same time, the prediction effect of a single section using the above method is acceptable, but if it is used to forecast the runoff of multiple sections in the watershed, the problem of water volume imbalance will often occur. However, the forecast of runoff in the final outlet basin is too high

In addition, the traditional mid- and long-term runoff forecasts have little consideration for the impact of reservoir regulation and storage. Even if there is consideration, it is simply divided into the flood limit water level in the flood season and the normal water storage level in the non-flood season. more complicated than this

Images