| Joints are widely exist in the rock mass. It is a complex geological structure, and has an important effect on the mechanics and permeability of rock mass. Due to the strength and stiffness of the jointed rock mass is much lower than intact rocks, so it’s important to systematic study on the geometry and mechanical properties and investigate the distribution of jointed rock mass. A large number of studies have shown that the geometric features of the joints are similar which have the same construction period and formation reasons. Due to the occurrence of joint surface has a great influence on the anisotropy and failure mode o f rock mass, it can use occurrence for partitioning of joints. Because of the complexity of the joint distribution and the limitations of the data collection, it is difficult to obtain the full distribution information of the joints, therefore, using mathe matical statistics method for grouping, and extracting the effective distribution information of joints will have great help on the further study of the rock mass. The thesis is mainly studying on the partitioning of joints, the basically research contents are as following:By establishing the mathematical model of the joints, selecting the reasonable index of distance which can be more really reflect the relation of distance between the joints. Then establishing the objective function which was as a basis for the quantitative grouping of the joints, and the grouping problem of joints will change into a simple mathematical extremum problem.To analysis and summary the existing grouping methods of joints, it mainly have the shortcomings such as the sensitive of isolated points and initial dividing of joints, it’s easy to fall into local optimum and the partitioning results are unstable. By introducing artificial bee colony algorithm in the thesis, making use of the advantages of the algorithm to improve the fuzzy C-means algorithm which based on the objective function, and it will obtain a more reasonable and stable result. Adopting the validity indexes to evaluate the superiority of the results of different groups when obtaining the results of the partitioning of joints, and then acquiring the optimal grouping result which will be more persuasive.Applying in random data of computer simulation and field survey data for partitioning of joints by the new algorithm, and obtaining the partitioning results of the joints respectively, then according to the validity indexes to select the final results. Comparing and analysis the final results and original results which are known of partitioning of the joints, and then verify the feasibility and practicability of the algorithm on partitioning of joints. |
