岩体结构均质区划分及岩体质量评价风险分析

发布时间：2018-03-19 08:32

本文选题：岩体质量　切入点：均质区划分　出处：《南京理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：工程岩体质量是影响高放废物地质处置工程长期稳定性和安全性的关键性因素,本文以甘肃北山预选区花岗岩岩体为研究对象,基于现场不连续面调查和实验资料,对北山芨芨槽块段各测区的岩体质量空间分布特征进行了研究。本文的主要研究内容与结论如下：(1)对原始不连续面采样数据进行了必要的前期预处理,然后基于Miller法原理,采用改进的91格大数目块网实现了对不满足Lancaster准则的情形的自适应分析与处理,从而得到了与实际地质条件及空间分布特征相吻合的均质区划分结果。(2)在总结已有产状分组聚类分析方法的基础上,对传统快速聚类算法进行了优化：根据最大密点原则对初始聚类中心进行了有效地确定；利用隶属度概念对产状矩阵进行了合理地优化；结合Xie-Beni指标对最优分组数进行了科学地判别；依据隶属度差异对随机离散裂隙进行了实时地剔除。以上优化算法进一步减小了聚类分析过程中可能存在的误差,从而保证了北山各测区产状优势分组结果的科学性与可靠性。(3)针对节理裂隙图像的特点,对传统Canny算法中的图像滤波、梯度计算、非极大值抑制及双闽值检测分别进行了优化,从而实现了对节理裂隙图像边缘信息的有效提取；在此基础上,根据量规法分维数计算原理,编程实现了对节理裂隙表面轮廓曲线分形维数的自动化计算,并建立了此分形维数与Barton粗糙度系数JRC间的量化关系。(4)以Q系统分级法为基础,结合北山地区的现场实测数据资料,对影响岩体质量的各主要参数进行了统计分析,并基于贝叶斯网络原理,利用Netica软件构建了相应的岩体质量评价风险分析模型,并对所建模型进行了应用和验证。
[Abstract]:The quality of engineering rock mass is the key factor to affect the long-term stability and safety of the geological disposal project of high-level radioactive waste. In this paper, the granite rock mass in Beishan preelection area of Gansu Province is taken as the research object, based on the field discontinuous surface investigation and experimental data. The spatial distribution characteristics of rock mass quality in each area of Beishan Jiji trough block are studied. The main contents and conclusions of this paper are as follows: 1) the necessary pre-processing of the original discontinuous surface sampling data is carried out, and then based on the principle of Miller method, An improved 91-lattice block network is used to realize adaptive analysis and processing for cases which do not satisfy the Lancaster criterion. Thus, the results of homogeneous area division consistent with the actual geological conditions and spatial distribution characteristics are obtained. Based on the conclusion of the existing cluster analysis methods of occurrence, The traditional fast clustering algorithm is optimized: the initial clustering center is effectively determined according to the maximum dense point principle, and the production matrix is reasonably optimized by using the concept of membership degree. Based on the Xie-Beni index, the optimal grouping number is scientifically judged, and the random discrete cracks are eliminated in real time according to the membership degree difference. The above optimization algorithm further reduces the possible errors in the clustering analysis process. Therefore, the scientific and reliability of the predominance grouping results of occurrence in the Beishan area are ensured. According to the characteristics of the joint fracture image, the image filtering, gradient calculation, non-maximum suppression and double threshold detection in the traditional Canny algorithm are optimized respectively. On the basis of this, according to the principle of fractal dimension calculation of gauge method, the automatic calculation of fractal dimension of joint fracture surface contour curve is realized by programming. The quantitative relationship between the fractal dimension and the Barton roughness coefficient (JRC) is established. Based on the Q system classification method and the field measured data in Beishan area, the main parameters affecting the rock mass quality are statistically analyzed. Based on the principle of Bayesian network, the corresponding risk analysis model of rock mass quality evaluation is constructed by using Netica software, and the model is applied and verified.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU45

【引证文献】