Automatically-calibrating rolling forecasting method for flood

Abstract

The invention relates to an automatically-calibrating rolling forecasting method for flood. The method is characterized by comprising the following steps of: acquiring a forecasting site determined by a user in a simulation area, and determining a topological relation between the forecasting site and upstream and downstream sites and a river reach according to a spatial topological relation of a river channel in the simulation area; extracting actually-measured water level or flow data of the along-way river channel in the simulation area at a simulation calculation moment, conducting interpolation on the actually-measured water level or flow data, and inputting the interpolated data into a verified rolling forecasting model for flood; outputting the forecast water level and flow data of the forecasting site according to the input data, the forecasting site and the upstream and downstream topological relation of the forecasting site through the rolling forecast model for flood; and carrying out data assimilation on the forecast water level and flow data by adopting measured data, wherein the rolling forecasting model for flood comprises a one-dimensional non-constant flow numerical model corresponding to the simulation area and an NAM hydrological model coupled with the one-dimensional non-constant flow numerical model. The method is applicable to the flood forecasting field.

Description

technical field [0001] The invention relates to an automatic calibrating flood rolling forecast method. Applicable to the field of flood forecasting. Background technique [0002] At present, the method of using real-time meteorological information data to drive numerical models for hydrological forecasting has been widely used, but most of the hydrological forecasts are only carried out based on traditional hydrological models. When using the model of coupled calculation of hydrology and hydrodynamics for real-time forecasting, the accuracy and timeliness of the forecast are often insufficient due to problems such as calculation stability and data fusion. The river channel model calculation method in the prior art has the following defects: [0003] 1) The stability of the hydrological-hydrodynamic coupling model is not high. When the hydrodynamic model is used for real-time calculation, it is often prone to instability problems such as divergence during model calculatio...

AI Technical Summary

Problems solved by technology

[0003] 1) The stability of hydrological and hydrodynamic coupling model is not high

When the hydrodynamic model is used for real-time calculation, it is often prone to instability problems such as divergence during model calculation due to complex river conditions and uneven quality of real-time data. These divergences will lead to model prediction. The accumulation of eventually results in an unusable

The one-dimensional river network model in the traditional forecasting method adopts the unified solution mode of circular and branched mixed river networks, and does not consider the existence of hydraulic structures in the river network, and the method does not consider the presence of dams, bridges, cofferdams, sluices, etc. in the river channel. It is impossible to accurately simulate the impact of hydraulic structures on the water level of the river, which has great limitations in practical engineering applications

[0004] 2) More and more real-time data are not effectively integrated and utilized

In recent years, with the continuous construction of hydrological monitoring stations, a large number of real-time rainfall stations and hydrological stations have been built and operated, and the massive available real-time data has not been effectively integrated and utilized in traditional forecasting methods

[0005] 3) The timeliness of forecast calculation is not enough

The current forecasting methods are based on traditional hydrological models and forecasting methods. When analyzing a large amount of data, effective data fusion technology is often not established, resulting in low time efficiency for processing real-time data, which often delays the efficiency of forecasting and leads to the forecast period. shorten

Images