: . 网络首发时间：2022-04-22 08:34:33 网络首发地址hnology, Kunming, Yunnan 650093, China; 2. School of Civil and Resources Engineering, University of Science and Technology Beijing, Beijing 100083, China)
Abstract: In response to the problems of the traditional single discontinuity clustering model, such as the risk of misclassification or omission and the inability to identify noise and orphan values, a method is proposed to group the discontinuity dominant partitioning of rock masses based on DBSCAN selective clustering ensemble. Firstly, the orientation of discontinuity is transformed into spatial coordinates, and the included angle sine of the unit normal vector is used as the similarity measure. Then, a certain number of different base clusters are constructed based on the DBSCAN algorithm, with the help of selective clustering ensemble technology, some excellent base clusters are selected. Finally, the consistent ensemble technology is used to fuse multiple base clustering results to obtain a highly reliable selective clustering ensemble result. The research results show that the clustering effect of this method is significantly higher than that of common clustering algorithms. The clustering results are objective and reasonable. It not only effectively identifies noise and solitary values, but also overcomes the shortcomings of easy over-segmentation or under-segmentation of a single model. The method is applied to the DIPS software data set and the discontinuity survey in the dam site area of Songta hydropower station to test the feasibility and effectiveness of the method. The research results have certain engineering value for accu
