Research On Real-Time Flood Forecasting In Large Basin

Flood forecasting is main dependence of flood controlling and decision, it is considered as an important non-structural measure on the flood control and mitigation . How to forecast its flood with high precision always is a difficult problem , for large basin's character of wide covering area and numerous branches. Real-time correct the large basin's forecasting result is one of the effective ways to improve the forecasting precision.The paper accomplish basin information collection, hydrology object's parse and basin's generalization on GIS platform, firstly, and on the base, it compartmentalizes large basin to different level subsystem, and compute from base to advance between each subsystem;It study basin's topology relation, solve the problem of object's best figure order according to the relation between each subsystem, construct project of flood forecasting. It compares the main actual real-time correction methods and find that adopt the method of error autoregression combine with attenuation memory least-squares can effectively correct flood forecasting's result. Lastly, it combine flood forecasting project with real-time correction model to put forward the project of flood forecasting's real-time correction for large basin.The study production of the paper is applied to hydrology can get good result, and flood forecasting's precision is much improved, so it is an applied method for flood forecasting 's real-time correction.