Back-fitting algorithm-based flood forecast real-time correction method

Abstract

The invention discloses a back-fitting algorithm-based flood forecast real-time correction method. According to the method, a back-fitting algorithm in a data mining technology is introduced based on a conventional error autoregression correction method to correct parameters of a hydrologic model, and stationary processing of a residual error sequence is considered, so that a new method for improving conventional flood forecast real-time correction is proposed. The method can be widely applied to real-time flood forecast, can effectively improve the flood forecast precision, and provides an important basis for a flood prevention dispatching decision.

Description

technical field [0001] The invention relates to the technical field of flood forecasting, in particular to a real-time correction method of error autoregression. Background technique [0002] Flood is one of the most serious natural disasters that threaten people's life and property in our country. Statistics show that the average annual loss caused by floods in my country ranks first in natural disasters. As a non-engineering measure in flood control and disaster reduction, flood forecast is an important basis for flood control decision-making. However, the hydrological model widely used in flood forecasting is derived from the inversion of measured rainfall and runoff data, which reflects the average condition of the underlying surface in the watershed, and there are still problems such as data error and model structure uncertainty. Causes unavoidable errors in flood forecasting. Therefore, it is a necessary measure to adopt real-time correction technology for proper cor...

AI Technical Summary

Problems solved by technology

[0011] In this process, there are also the following problems: (1) The error autoregressive method often ignores the applicable conditions of the AR model when using the AR model to fit the residual sequence, that is, the residual sequence is required to be a stationary sequence; (2) combined with the error autoregressive In the real-time flood forecasting method of the regression correction method, most of the observed historical hydrological data are used, and the parameters of the hydrological model are determined first. The matching degree of the measured flow rate may not be optimal, so there is still room for improving the forecast accuracy

[0012] In the literature "Neural networks and non-parametric methods for improving realtime flood forecasting through conceptual hydrological models" (Brath et al., 2002), it is mentioned that the non-stationary residual series is transformed into a residual series by difference; but for the traditional AR model "first Determine the parameters of the hydrological model, and then carry out the residual correction" This kind of improvement is rare

Images