Urban flood model runoff sensitive parameter identification method based on artificial neural network

Abstract

The invention provides an urban flood model runoff sensitive parameter identification method based on an artificial neural network. The method comprises the following steps: dividing a city into a plurality of hydrological response units according to urban runoff characteristics; and by taking the environmental indexes influencing the runoff parameter sensitivity of the hydrological response unit as input and the sensitive parameters of the rainstorm flood management model as output, identifying the sensitive runoff parameters of the rainstorm flood management model by using an artificial neural network. According to the method, the complex process of a traditional method can be skipped, and the sensitive parameters can be obtained only by inputting the environment indexes, so that the possibility is provided for real-time correction of the urban flood model.

Description

technical field [0001] The invention relates to the technical field of parameter sensitivity analysis of an urban flood model, in particular to an identification method for runoff sensitive parameters of an urban flood model based on an artificial neural network. Background technique [0002] Urban flooding has become an important challenge restricting urban development, and effective rainwater management using urban flood models is an effective means to deal with this challenge. As the basis of model calibration, analyzing the sensitivity of model parameters is of great significance for improving the accuracy of model simulation. Sensitivity analysis of urban flood model parameters is an important step in urban flood simulation and model parameter calibration. Efficient and accurate acquisition of sensitive parameters is the key to real-time model calibration and flood dynamic simulation. [0003] At present, there have been a large number of studies on the sensitivity an...

AI Technical Summary

Problems solved by technology

Although parameter sensitivity analysis can reduce the number of parameters that need to be calibrated, optimizing the parameters of the entire model requires a lot of calculations, and it is very difficult to optimize parameters quickly

[0005] The traditional parameter sensitivity analysis method is complicated, and requires multiple adjustments and simulations of the model, which limits the speed of sensitive parameter acquisition and the efficiency of model correction, and hinders the realization of real-time correction of urban flood models and dynamic prediction of flood processes

Images