基于典型小概率法-LSTM的大壩預(yù)警技術(shù)研究
打開文本圖片集
關(guān)鍵詞：大壩安全；典型小概率法；LSTM；監(jiān)測預(yù)警；三門峽水庫中圖分類號：TV87 文獻標(biāo)志碼：A doi：10.3969/j.issn.1000-1379.2025.07.013引用格式：，，，等.基于典型小概率法-LSTM的大壩預(yù)警技術(shù)研究[J].人民黃河，2025，47（7）：78-83.
Research on Dam Early Warning Technology Based on Typical Small Probability Method-LSTM
LI Mingyang 1，2，3 ， DENG Yu 1，3 ， ZHANG Baosen 1，3 ， GUO Jinjun2 （1.Yellw River Institute of Hydraulic Research， YRCC， Zhengzhou 450oO3，China; 2.School of Water Conservancy and Transportation， Zhengzhou University，Zhengzhou 45ooo1，China; 3.Research Center on Levee Safety and Disaster Prevention，MWR，Zhengzhou 45OOO3，China）
Abstract：Teperatioalsusofteameniaeseiitcallypacswatersuplsecuritdfdprevetofdowsai iesalong eYelowRiverHowever，onventioaloitoigapproacsrelingonssortwoksandmanualbservatiossuferfroin adequatereal-tpefoaedisuentpreTdsselatios，issdeveedtegadrld elincorporatingulti-soucedataanddamicesponsemechanisms.Themethodologinvoledcalulatingwaingtresholdsforrialprameters（e.g，damdisplacementandsepagepresure）usingtheTypicalSmallProbabilityMethod，coupledwithanLSTM-basedpredic tionmodelforficient waringtoughthresholdcomparisonValidationusingtheSanmenxiaReservoircasedemonstratesthemodel’ssuperior predictive accuracy compared to traditional methods.
KeyWords：dam safety；typical smallprobabilitymethod;LSTM；monitoring and early warning；Sanmenxia Reservoir
0 引言
黃河三門峽水庫作為黃河流域防洪與水資源調(diào)配的重要樞紐，其安全運行直接影響下游城市供水與生態(tài)穩(wěn)定，不僅是黃河流域水資源調(diào)配和防洪減災(zāi)的樞紐工程，也是保障周邊地區(qū)經(jīng)濟社會穩(wěn)定發(fā)展的生命線[1-3]。（剩余8227字）