基于降雨预报滚动修正和模型参数实时校正的城市洪涝动态预报研究OA

To address the practical demand for high-precision dynamic flood forecasting in urban flood risk prevention and control,a dynamic urban flood forecasting method based on rolling correction of rainfall forecasts and real-time calibration of model parameters is proposed,in which,the measured rainfall data are used to perform rolling correction on real-time collected rainfall forecast data,and the rolling-corrected rainfall forecast data are then used as input for the urban flood model.During the flood process,the model parameters are calibrated in real time based on measured errors to drive the dynamic update of flood waterlogging process forecast results,thereby achieving dynamic improvement of flood forecasting accuracy throughout the rainfall-flood process.Taking the new campus of Zhengzhou University as the study area,the results show that the forecasting accuracy of verified waterlogging points improves dynamically over time,with the Nash-Sutcliffe efficiency coefficient(NSE)gradually increasing from less than 0.6 to 0.95.This indicates that the dynamic urban flood forecasting method can effectively improve forecasting accuracy and meets current practical needs of urban flood prevention and control in China.