大渡河流域汛期涨水阶段径流相似性预报  

Runoff similarity forecast during water rise stage of flood season in the Dadu River Basin

作　　者：谭乔凤 施颖 闻昕[1] 杨杉 TAN Qiaofeng;SHI Ying;WEN Xin;YANG Shan(College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China)

机构地区：[1]河海大学水利水电学院,江苏南京210098

出　　处：《水资源保护》2024年第6期139-147,共9页Water Resources Protection

基　　金：国家重点研发计划项目(2019YFE0105200)。

摘　　要：以大渡河流域干流上游丹巴断面为例,对流域涨水规律进行全面分析,结合产汇流特征和下游电站调度需求,利用遍历组合寻优方法确定了涨水模式分类阈值和涨水判断条件,提出了基于“形值”相似的分类别涨水径流预报方案,实现了预见期为10 d的径流逐日滚动预报。结果表明:2019—2021年,大渡河流域汛期大雨、中雨、小雨的涨水识别准确率分别达到100%、93.18%、84.81%;相比于不考虑涨水条件识别的基础预报方案,分类别涨水预报方案在不同涨水模式下的径流预报平均相对误差降低了0.45%~6.81%;预见期为1、5、10 d的径流预报平均相对误差分别为3.39%、9.64%、12.29%,纳什效率系数分别达到0.99、0.95、0.91,该方法具有较高的预报精度。Taking the Danba Section upstream of the main stream of the Dadu River Basin as an example,a comprehensive analysis of the water rise law in the basin was conducted.Combining the characteristics of runoff and the scheduling needs of downstream power stations,the threshold for water rise mode classification and the conditions for water rise judgment were determined using the traversal combination optimization method.A classified water rise runoff forecasting scheme based on“form-value”similarity was proposed,achieving a daily rolling forecast of runoff with a prediction period of 10 d.The results showed that from 2019 to 2021,the accuracy of identifying rising water in heavy rain,moderate rain,and light rain during the flood season in the Dadu River Basin reached 100%,93.18%,and 84.81%,respectively.Compared with the basic forecasting scheme that does not consider the identification of water rise conditions,the average relative error of runoff forecasting for different water rise modes is reduced by 0.45% to 6.81% for the classified water rise runoff forecasting scheme.The average relative errors of runoff forecasts with forecast periods of 1,5,and 10 d are 3.39%,9.64%,12.29%,respectively,and the Nash efficiency coefficients reach 0.99,0.95,and 0.91,respectively.This method has high forecast accuracy.

关 键 词：涨水识别 相似性预报 径流滚动预报 大渡河流域 

分 类 号：P338[天文地球—水文科学]