龙滩水电站相似性径流预报研究  

Study on Similarity Runoff Forecasting of Longtan Hydropower Station

作　　者：谢金记 陈粉娥 杨杉 李凌燕 钟成 闻昕[2] XIE Jin-ji;CHEN Fen-e;YANG Shan;LI Ling-yan;ZHONG Cheng;WEN Xin(Guangxi Guiguan Electric Power Co.,Ltd.,Nanning 530000,China;College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China)

机构地区：[1]广西桂冠电力股份有限公司,广西南宁530000 [2]河海大学水利水电学院,江苏南京210098

出　　处：《水电能源科学》2023年第11期10-13,共4页Water Resources and Power

摘　　要：针对岩溶地区产汇流机制复杂、地下水监测资料不足、径流预报精度低等问题,提出了融合物理机制和数据驱动的相似性预测方法,建立了不同流量量级、不同预报因子、不同前期影响滞时下最优组合预报方案,可实现不同水雨情条件下多模式的自适应匹配和智能切换,可生成具有可解释性的径流预报结果,并将该方法应用于龙滩水电站径流预报中。结果表明,该方法提高了岩溶地区的径流精度和有效预见期,对保障大型电站科学调度和安全运行具有科学意义和实际价值。In response to the complex mechanism of runoff yield and concentration,insufficient groundwater monitoring data,and low accuracy of runoff prediction in Karst areas,a similarity prediction method that integrates physical mechanisms and data-driven approaches has been proposed.The optimal combination prediction scheme has been established for different flow magnitudes,prediction factors,and preceding affected lag-time,which can achieve adaptive matching and intelligent switching of multiple modes under different water and rain conditions.It can generate interpretable runoff prediction results.This method is applied to the runoff prediction of Longtan Hydropower Station.The results indicate that this method improves the accuracy and effective prediction period of runoff in Karst areas,which has scientific significance and practical value for ensuring the scientific scheduling and safe operation of large power plants.

关 键 词：相似性识别 最近邻抽样 滚动预报 岩溶 多模式切换 

分 类 号：TV743[水利工程—水利水电工程] P338[天文地球—水文科学]