基于小样本及区间数据的腐蚀PC梁抗力概率分析

发布时间：2018-01-10 00:16

本文关键词：基于小样本及区间数据的腐蚀PC梁抗力概率分析　出处：《长沙理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：在我国现有预应力混凝土桥梁中,被各种不利因素腐蚀的PC桥梁逐年在增加,尤其是基于氯盐环境下。考虑这种环境因素下,由于预应力钢绞线的腐蚀从而导致梁的抗力衰减,严重影响了桥梁在服役时的安全性。并在此基础上,根据似然函数和最大熵理论,建立了由两类数据(小样本和区间数据)来源引起的不确定性PC桥梁抗力退化分析方法。文章主要是从以下几个方面展开研究:(1)介绍了在氯盐环境下由于钢绞线的腐蚀导致PC桥梁抗力退化的研究现状,并分析目前国内外在服役期间既有PC桥梁结构面临的问题和不确定性的研究现状,还对钢绞线的腐蚀影响因素进行了分析讨论。(2)在服役PC桥梁中,由于主观因素和客观因素等条件的影响,会产生各类无法预知的不确定性。文章介绍了PC桥梁在服役过程中不确定性的来源,如何将这些不确定性进行分离和不确定性量化做了相关研究。本文利用似然函数和最大熵理论来处理由两类数据来源引起的不确定性,这比不考虑不确定时(这种情况下是用最大似然估计法来计算的)更为可靠。(3)由于钢绞线在化学成分和组织结构的不同,使得钢绞线在不利环境下的腐蚀与普通钢筋的腐蚀有所不同。本文主要工作如下:首先,对普通钢筋和钢绞线在不利环境下的腐蚀形态特点进行了比较分析;对于钢绞线锈蚀主要后蚀坑的形状进行了讨论,并且分析和总结了蚀坑的深度、长度和宽度的分布特征,最后利用似然函数的方法来讨论基于小样本数据不确定性条件下蚀坑的最大深度和最大宽度的PDF(概率密度函数)。(4)在不利环境下,结合国内外学者已经推导的腐蚀后钢绞线的承载力模型,并且在计算PC梁的承载力模型时引入变形协调系数能够得到经氯离子腐蚀以后PC梁的抗弯承载力模型。然后以一座受弯实桥为例,分析了在两类数据不确定性条件下对抗力的影响。结果表明:以似然函数和最大熵理论来构造变量的概率密度函数,能够减少主观不确定性,由此得到的结果更加接近真实情况。
[Abstract]:In the existing prestressed concrete bridges in China, the number of PC bridges corroded by various adverse factors is increasing year by year, especially under the chloride environment. The corrosion of prestressed steel strands leads to the attenuation of the beam resistance, which seriously affects the safety of the bridge in service. On this basis, according to the likelihood function and maximum entropy theory. A method for analyzing the resistance degradation of PC bridges with uncertainty caused by two kinds of data (small sample and interval data) is established. The research status of PC bridge resistance degradation due to the corrosion of steel strands in chloride environment is introduced. The problems and uncertainties of existing PC bridges in service at home and abroad are analyzed, and the factors affecting the corrosion of steel strands are analyzed and discussed. Due to the influence of subjective factors and objective factors, various kinds of unpredictable uncertainties can be produced. This paper introduces the sources of uncertainty in the service of PC bridges. How to separate and quantify these uncertainties is studied. In this paper, the likelihood function and maximum entropy theory are used to deal with the uncertainty caused by two kinds of data sources. This is more reliable than when uncertainty is not taken into account (in this case the maximum likelihood estimation method is used) because of the different chemical composition and structure of the steel strands. The main work of this paper is as follows: firstly, the corrosion characteristics of common steel bar and steel strand in adverse environment are compared and analyzed; The shape of the main corrosion pits is discussed, and the distribution characteristics of the depth, length and width of the corrosion pits are analyzed and summarized. Finally, using the method of likelihood function, the PDF (probability density function) (PDF) of the maximum depth and width of the etch pits under the uncertainty of small sample data is discussed in the unfavorable environment. Combined with the domestic and foreign scholars have deduced the corrosion of steel strand bearing capacity model. When calculating the bearing capacity model of PC beam, the bending bearing capacity model of PC beam after chloride ion corrosion can be obtained by introducing the deformation coordination coefficient. Then, a real flexural bridge is taken as an example. The effect of antagonistic force on two kinds of data uncertainty is analyzed. The results show that the probability density function of variables can be reduced by using likelihood function and maximum entropy theory. The result is closer to reality.
【学位授予单位】：长沙理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U441;U448.35

【参考文献】