篮家岩隧道洞口段围岩参数反分析及施工数值模拟研究

发布时间：2018-07-14 17:57

【摘要】：随着国民经济的快速发展,我国修建了大量的铁路和公路,隧道等地下结构建设面临着巨大的机遇与挑战。地下工程所赋存地质体的物理力学参数对隧道工程的设计、施工和力学分析有着重要意义,关系到地下工程的安全性和经济性。然而由于地下环境的复杂性,使得这些重要的物理力学参数难以确定。为解决这一问题,本文以篮家岩隧道洞口段为研究对象,研究了基于MEABP神经网络的位移反分析方法并将反演得到的围岩参数用于主洞隧道施工数值模拟,分析了隧道施工中围岩-支护结构位移和应力特点,研究了隧道变形的全过程曲线和隧道洞壁各测点位移释放率随时间和位置的关系。本文主要研究内容和成果有以下几个方面:(1)对平行导洞监测数据进行回归分析,并对隧道监测数据的稳定值进行预测。在隧道开挖空间效应的基础上,根据前人研究,参考山东大学李术才对Ⅴ级围岩上下台阶法开挖后释放率计算值,以平导PDK51+326断面为例计算了隧道测点的真实变形量。(2)针对BP神经网络的缺陷采用思维进化算法优化BP网络,根据正交试验方案采用FLAC2D对平导洞正算构建位移反分析样本集,通过学习训练建立基于MEABP神经网络的位移反分析系统,并输入计算的真实变形量反演出该段隧道的围岩参数。(3)根据平行导洞反演的围岩参数正算主洞隧道施工,并对K51+310断面在上下台阶开挖后的位移和应力分析,得出了隧道施工中该断面的位移和应力分布规律。(4)根据主洞施工中拱顶、拱腰、拱脚以及它们中间部位的水平和竖向位移全过程曲线变化,分析了其变化趋势及原因,发现上下台阶法施工的隧道,洞壁各关键部位竖向位移变化受下台阶开挖影响比水平位移受的影响小。(5)根据数值模拟中对K51+326断面的记录,研究了隧道洞壁各关键部位(拱顶、拱腰、拱脚和它们的中间部位)的位移和位移释放率随时间的变化曲线,得出了以下结论:靠近拱顶部位的位移释放率普遍偏大,沿着洞壁从拱顶到拱脚位移和位移释逐渐减小,拱脚处达到最小;测点位置固定时,测点位移及释放率随时间大致符合S曲线;时间固定时,各关键部位测点位移和位移释放率大致关于测点位置与竖向的夹角呈二次抛物线关系;由于研究问题的特殊性,将位移释放率函数简化为关于测点位置与竖向夹角?和时间t的二元函数,采用多元回归得出表达式。对比拟合误差,发现该函数在t?0附近区域有少量误差,在t?0区域且稍远离t?0的区域误差较小,回归效果较好。(6)最后根据已研究的位移释放率的变化规律,结合K51+326断面实测拱顶沉降序列,采用S型曲线对实测序列进行修正,得出了隧道施工断面变形的全过程曲线表达式。
[Abstract]:With the rapid development of national economy, China has built a large number of railways, highways, tunnels and other underground structure construction is facing enormous opportunities and challenges. The physical and mechanical parameters of geological bodies in underground engineering are of great significance to the design, construction and mechanical analysis of tunnel engineering, and are related to the safety and economy of underground engineering. However, due to the complexity of underground environment, these important physical and mechanical parameters are difficult to determine. In order to solve this problem, the displacement back analysis method based on MEABP neural network is studied in this paper, and the parameters of surrounding rock are used to simulate the construction of main tunnel. The characteristics of displacement and stress of surrounding rock support structure in tunnel construction are analyzed. The relationship between displacement release rate and time and position of tunnel wall is studied. The main contents and results of this paper are as follows: (1) regression analysis of monitoring data of parallel guide holes and prediction of the stability of tunnel monitoring data. On the basis of the spatial effect of tunnel excavation, according to the previous studies and referring to Li Gongcai of Shandong University, we calculated the release rate of class V surrounding rock after excavation with the upper and lower step method. Taking section PDK51 326 as an example, the true deformation of tunnel measuring points is calculated. (2) to solve the defects of BP neural network, the thought evolution algorithm is used to optimize BP network. According to the orthogonal test scheme, FLAC2D is used to construct the sample set of displacement and back analysis. The displacement back analysis system based on MEABP neural network is established by learning and training, and the parameters of surrounding rock of this section of tunnel are inputted and calculated. (3) the construction of main tunnel is directly calculated according to the parameters of surrounding rock inversed by parallel guide tunnel. Based on the analysis of displacement and stress of K51 310 section after excavation of upper and lower steps, the law of displacement and stress distribution of this section in tunnel construction is obtained. (4) according to the arch top and waist in the construction of main tunnel, The whole process curve of horizontal and vertical displacement of arch foot and its middle part is changed. The change trend and reason are analyzed, and the tunnel constructed by upper and lower step method is found. The variation of vertical displacement in every key part of tunnel wall is less affected by lower step excavation than horizontal displacement. (5) according to the records of K51 326 section in numerical simulation, the key parts of tunnel wall (arch roof, arch waist) are studied. The displacement and displacement release rate of arch foot and their middle part) change with time. The following conclusions are drawn: the displacement release rate near arch top is generally larger, and the displacement and displacement release decrease gradually along the wall from arch top to arch foot. When the measuring point position is fixed, the displacement and release rate of the measuring point approximately accord with S curve with time, and when the time is fixed, The displacement and displacement release rate of each key position is about the quadratic parabola relation between the measuring point position and the vertical angle, because of the particularity of the research problem, the displacement release rate function is simplified to be about the measuring point position and the vertical angle? And the binary function of time t, the expression is obtained by multivariate regression. By comparing the fitting error, it is found that the function has a little error in the region near t0, a small error in the region of t0 and a little distance from t0, and a good regression effect. (6) finally, according to the variation of displacement release rate, Combined with the measured dome settlement sequence of section K51 326, the S-shaped curve is used to modify the measured sequence, and the whole process curve expression of section deformation in tunnel construction is obtained.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U452.12;U455.4

【相似文献】