摘要
特高拱坝运行中受多因素协同作用,不同高程和区域的变形规律差异较大,坝体局部单测点的变形量并不能代表大坝整体的安全性发生变化。为此,综合考虑多测点、多测次监测量的内在关联性,构建了特高拱坝变形面板系列的表征方法;基于面板数据聚类理论,建立了大坝变形相似性指标及度量方法;利用熵权法构建大坝变形区域数的确定准则,提出了特高拱坝变形分区方法及实现步骤;通过利用特高拱坝两个维度的变形监测信息,将传统的"点"分析方法转变为区域分析方法。某特高抛物线双曲拱坝实例应用表明,该方法可根据39个变形监测点确定六大类测点,每类测点可以综合描述大坝对应区域的总体变形特征,再根据测点所在的位置将大坝分为六个变形区域,有效描述了特高拱坝的实际变形规律,可弥补传统分析方法上的一些不足,为特高拱坝变形分析提供了新方法。
Owing to high arc dam operation affected by many factors,there are large differences of the deformation law in different elevation and region of the dam.Dam deformation in local single measuring point cannot represent the change of the whole dam safety.Firstly,internal relevance among multi-measuring points and multiple measurements is considered to express high arch dam deformation series using panel data model.Then,based on the clustering theory of panel data,similarity index and method for measuring the dam deformation are established.Besides,entropy weight method is used to build the standard of dam deformation region numbers,and partitioning method and implementation steps of high arch dam deformation are put forward.By using two dimensions of high arch dam deformation monitoring information,the traditional‘point＇analysis method is transferred into regional analysis method.One high parabolic hyperbolic arch dam application shows that this method can determine six categories of measuring points based on thirty nine deformation points.Each category can comprehensively describe the whole deformation features of the relevant areas in dam.The dam is divided into six deformation areas according to the positions of measuring points,which effectively describes the real deformation rule of high arch dam.This method can make up the deficiency of the traditional analysis method and provides a new method for deformation analysis of high arch dam.
出处
《水电能源科学》
北大核心
2016年第10期40-43,18,共5页
Water Resources and Power
基金
国家自然科学基金重点项目(51139001
41323001)
国家自然科学基金项目(51479054
51579086
51379068
51579083)
关键词
特高拱坝
变形性态
面板数据
变形分区方法
high arch dam
deformation state
panel data
deformation partitioning method
