基于激光扫描技术的岩坡工程精细化块体识别方法研究

The complex block system is formed by the interaction between the free face and discontinuities in rock slope engineering. The accuracy of the block morphology is determined by the refined degree of the rock mass model. However, it is the bottleneck of block theory that they could not combine with each other due to the limit of the block identification method. This study aims to the block identification in the refined model of rock mass. Firstly, TIN is built to characterize geometrical morphology of free face accurately based on point cloud data from three-dimensional scanning technique. Secondly, random discontinuities is identified form point cloud coupled with photogrammetric technology, and then the network model of random discontinuities is built with the method of Monte Carlo. Fixed discontinuities are built by TIN base on characteristic of curved face captured by the borehole camera technology. Thirdly, the model of the free face is decomposed into convex polyhedrons by projection method, and then convex polyhedrons are dispersed into small block elements. Incision of small block elements by surface discontinues is realized according to the principle of closure. On the basis, spatial block identification between two types of discontinues is realized using the results of the previous research. Finally, identification methods of discontinues and blocks are verified by the model experiments. This study combines the refined characterization of rock mass structure and the block theory, which provides development space for the block theory. It also provides the generation algorithm of block unit for the discontinuous rock mass mechanics.

在岩质边坡工程中，临空面与结构面相互作用形成了复杂的块体系统，块体形态的准确度由地质模型的精细程度决定，但由于块体识别方法的限制，二者的结合仍是制约块体理论发展的瓶颈。本研究以精细化岩坡模型中全空间块体的识别为目标，首先，基于三维激光扫描技术获取岩坡临空面点云数据，构建三角面单元网格，实现临空面形态的精细化表征。其次，辅以摄影测量技术从激光点云中识别出随机结构面，进而利用蒙特卡罗法建立三维网络模型。利用钻孔摄像技术获取确定性结构面的曲面特征，并以三角面单元构建。再次，采用投影法实现临空面模型的凸化分解并进行单元离散，根据多面体的封闭性实现曲面结构面对单元块体的切割，在此基础上利用前期研究成果实现两类结构面间的全空间块体识别。最后，制作室内模型对结构面及块体的识别方法进行验证。本研究将岩体结构精细化表征与块体识别相结合，为块体理论提供发展空间，同时也为非连续岩体力学提供块体生成算法。

随着国家基础建设的快速发展，特别是西部大开发的实施，位于山区的水利水电、交通市政、露天矿山开采等工程与日俱增，大量人工边坡不断成为工程建设的重大危险源。岩质边坡失稳破坏带来的损失较为严重，本研究以精细化岩坡模型中全空间块体的识别为目标，首先，基于三维激光扫描技术获取岩坡临空面点云数据，构建了三角面单元网格，实现了临空面形态的精细化表征。其次，辅以摄影测量技术从激光点云中识别出随机结构面，进而利用蒙特卡罗法建立三维网络模型。利用钻孔摄像技术获取确定性结构面的曲面特征，完成三角面单元构建。再次，采用投影法实现临空面模型的凸化分解并进行单元离散，根据多面体的封闭性实现曲面结构面对单元块体的切割，在此基础上利用前期研究成果实现两类结构面间的全空间块体识别。最后制作室内模型对结构面及块体的识别方法进行了验证。本次研究中将岩体结构精细化表征与块体识别相结合，为块体理论提供发展空间，同时也为非连续岩体力学提供块体生成算法。同时，三维激光扫描技术可以较好的应用于岩质边坡工程中，实现地质信息采集的自动化，充分发挥三维激光扫描技术精度高、量测速度快、非接触式测量等优点，服务于国家基础建设，具有学科交叉渗透的深远影响和极其重要的现实意义。. 依托课题研究，目前已发表学术论文7篇，其中包含1篇专业顶级期刊论文（发表于Computers and Geotechnics），申请专利5项，授权发明专利1项，实用新型专利4项。团队成员多次参加学术交流，培养研究生多名。