Topological And Statistical Study On Grouping Of Fractures For Jointed Rock Mass

Rock mass is cut by one or more groups of parallel weak structural planes. Therefore, rock mass is more often referred as the fractured rock mass. Fractures are fissures or small faults in the fractured rock mass and when the rock is fractured by forces, the blocks on both sides show little relative displacement.The physical and geometrical characteristic of the structural planes is an important factor which affects the permeability and mechanical properties of the rock mass. The number and direction of fracture group are important parts of the geometric properties of the fractures, and the statistical analysis and grouping for the occurrence of fractured rock is a basic study of rock mechanics. Because of the complexity of the fracture distribution and the limitations of available data, it is difficult to obtain detailed fracture data. Data processing and analysis are required to extract the effective information of the fractures. So far, it is still lack of effective processing method for discrete data processing and analysis.In this research, the distribution of the fractures in the rock mass of a mine is studied, and the procedures of the fracture grouping are refined. The discrete data is optimized to make the information extraction be without the interference of the random cracks. And the fracture data is processed to get more reasonable fracture group and effective information to reflect the fracture characteristics.Specific procedures are as followed:(1) The fracture data was collected by site investigation in the mine area, and the consistency and randomness of the occurrence of the fractures is studied.(2) With the tool of Matlab, the topological analysis for the fracture data is implemented, including data conversion, data mapping and data analysis, in the coordinate system of stereographic projection, and the qualitative results of the fracture grouping are obtained.(3) According to the principles of statistical analysis, the fractures are grouped several times by using the method of the fast (k-means) cluster, and the optimal grouping was chosen according to the dispersion coefficient from the function of the fisher distribution, and the preliminary quantitative results of the fracture grouping are obtained.(4) By using the method of the fast cluster with super multi-cluster and radius scanning cluster, the random cracks are deleted from the fracture data, and the effective information about the fractures is obtained. In this research, the fractures are grouped by using several methods, and the procedures of the fracture grouping are extracted; and the discrete fracture data is processed effectively to make sure the random cracks cannot affect the results of the facture grouping, and the accurate information can be obtained. Compared with the conventional procedures, the procedures improve the optimization of discrete data, which makes the scope of the fracture grouping larger and the reliability increase.