基于力学等效的岩体关键节理迹长阈值研究

Study on the threshold of key joint trace length in rock mass based on mechanical equivalence

为了在复杂节理网络中获得关键节理,为结构面选择性测量提供科学定量标准,实现节理网络的简化,以迹长为变量研究关键节理迹长阈值。采用3D打印技术,并根据Monte Carlo理论制备随机节理网络模型,浇筑成100 mm×20 mm×100 mm平板试样,在单轴压缩作用下对类岩体完整性系数、强度、变形和能量特性进行了分析,基于节理岩体力学等效原理,研究确定关键节理的迹长阈值。研究结果表明:(1)节理迹长是影响试样力学特性的核心要素之一。当剔除小于某一长度节理时,试样强度、变形和能量特性均无显著变化。(2)基于主成分分析和高斯混合模型聚类算法,建立节理岩体力学等效分析方法,实现鉴于多种力学参数岩体质量分类,进而对比力学不等效试样节理网络差异,形成一套关键节理迹长阈值研究方法。(3)综合考虑9种物理力学参数,认为试样边长的19%为关键节理的迹长阈值,对试样力学特性影响显著,为未来人工智能分析节理岩体工程如何科学取舍节理提供理论依据。

In order to select the key joints from the complex joint network,to provide a scientific quantitative criteria for the selective measurement of structural planes and to simplify the joint network,the threshold value of key joint trace length was studied by taking the trace length as a variable. The random joint network models were prepared by 3D printing technology and based on Monte Carlo theory,and flat specimens of 100 mm ×20 mm ×100mm were cast. The integrity coefficient,strength,deformation and energy properties of jointed rock mass were analyzed under uniaxial compression, and the threshold value of the key joint trace length was studied based on the mechanical equivalence principle of jointed rock mass. The results show that:(1) The joint trace length is one of the key factors affecting the mechanical properties of specimens. The strength,deformation and energy properties of the specimens do not change significantly when the joints smaller than a certain length are removed.(2) Based on the principal component analysis and Gaussian mixed model clustering algorithm,a mechanical equivalence analysis method of jointed rock masses was established to realize the classification of rock masses with various mechanical parameters,and the differences of the joint network of mechanically unequal samples were analyzed. A set of key joint trace length threshold research method was formed.(3) Considering 9 physical and mechanical parameters,19% of the specimen edge length is considered as the trace length threshold of key joints which significantly impacts the mechanical properties of the specimens. The research results provide a theoretical basis for future artificial intelligence analysis of how to scientifically select joints in jointed rock projects.