浙江省无资料流域洪水预报方法研究

Flood Prediction in Ungauged Basins of Zhejiang Province

无资料或资料不足是任意流域开展洪水预报面,临的难题和挑战。在浙江全省适用水文预报模型筛选基础上,利用自动寻优技术对浙江省39个代表流域978场暴雨洪水进行逐流域、逐场洪水产、汇流参数率定,并根据参数变幅、敏感性,分别采用均值、等值线空间内插、结合流域特征运用投影寻踪回归分析等方法进行产、汇流参数的地区综合,提出了全省无资料流域洪水预报方法,并经预报模型检验,精度可以满足防洪预警要求。

Flood forecasting in any area with no data or insufficient data is a problem and challenge. In this paper, on the basis of screen of hydrologic forecast models which are suitable for the whole province, the runoff yield and flow concentration parameters of 978 storm floods in 39 representative basins in Zhejiang Province were calibrated using automatic optimization technology one by one. According to the parameter range and sensitivity, methods as average and contour spatial interpolation, regression analysis using projection pursuit combined with the characteristics of river basins were used to realize regional comprehensive of the runoff yield and flow concentration parameters. The flood forecasting methods for the ungauged areas in the whole province have been proposed and examined by predictiou models, and the precision can meet the requirements of flood warning.