發(fā)布時間：2019-02-21 17:33

【摘要】：將極值理論中的廣義Pareto分布模型用于大壩預(yù)警指標(biāo)擬定中是一個行之有效的方法,該方法通過設(shè)定某一閾值,以監(jiān)測序列超過該閾值的實測值為分析樣本,并認(rèn)為該樣本服從廣義Pareto分布。由于閾值的選取關(guān)系到廣義Pareto分布參數(shù)估計的精度,而用超限期望圖法確定的閾值具有較大的主觀性誤差,因此,從定義出發(fā),通過數(shù)學(xué)推導(dǎo),建立了自動選取閾值的方法。通過實例分析表明,采用自動選取法確定閾值的廣義Pareto分布擬定的大壩預(yù)警指標(biāo)更加合理。
[Abstract]:It is an effective method to apply the generalized Pareto distribution model in extreme value theory to the determination of dam warning index. By setting a certain threshold value, the measured value of monitoring sequence exceeding that threshold is taken as the analysis sample. It is considered that the sample is based on the generalized Pareto distribution. Because the selection of threshold value is related to the accuracy of parameter estimation of generalized Pareto distribution, and the threshold value determined by the method of over-limit expectation graph has great subjective error, a method of automatically selecting threshold value is established from the definition and mathematical derivation. The analysis of examples shows that the dam warning index based on the generalized Pareto distribution with automatic selection method is more reasonable.
【作者單位】： 西京學(xué)院土木工程學(xué)院;
【基金】：陜西省教育廳科學(xué)研究項目(15JK2171)
【分類號】：TV698.1

【相似文獻(xiàn)】

相關(guān)期刊論文 前6條

1 王劍峰;宋松柏;;廣義Pareto分布在超定量洪水序列頻率分析中的應(yīng)用[J];西北農(nóng)林科技大學(xué)學(xué)報(自然科學(xué)版);2010年02期

2 陳子q，

本文編號：2427725