文献类型：期刊文献

英文题名：Rapid Identification and Efficient Resolution of Rock Mass Structural Planes Using the Opff Algorithm: Applications in Slope Stability Analysis

作者：Guan, Shenggong[1]; Yuan, Mingsong[1]; Wu, Faquan[1]; Nan, Hu[1]; Zheng, Hongchao[2]

机构：[1] International Joint Laboratory for Dijital Geotechniques and Intelligent Exploration, Shaoxing University, Zhejiang Province, China; [2] Faculty of Engineering, China, University of Geosciences·[Wuhan], Wuhan, China

年份：2025

外文期刊名：SSRN

收录：EI(收录号：20250211316)

语种：英文

外文关键词：Landslides

外文摘要：In large-scale engineering projects, structural planes in rock masses, whether natural or externally induced, are vital for slope stability. As rock mass engineering projects become more common, there is a growing need to identify these planes. This study proposes a novel OPFF algorithm. This algorithm integrates the Octree algorithm, Principal Component Analysis (PCA), Firefly algorithm, and Fuzzy C-Means algorithm to achieve rapid identification and efficient computation of rock structure planes. The structural surface recognition algorithm first processes the point clouds data of the rock mass. It segments the data using the Octree algorithm and determines the structural planes through Principal Component Analysis. Subsequently, it clusters normal vectors for structural planes using the Firefly algorithm. To enhance the clustering accuracy, the discontinuous sets of rock mass structural planes are further refined using Fuzzy C-Means clustering. This clustering method determines the optimal cluster centers, thus facilitating the efficient identification of rock mass structural surfaces. Applied to datasets from Ouray, Colorado in the US and the Eagle Mountain slope in Shaoxing, China, the OPFF algorithm can quickly obtain information like the direction, dip angle, and spacing of rock mass structural planes. The error between its output and manual measurement is less than 5%, showing high accuracy. Compared to other algorithms, it has higher computational efficiency in structural plane recognition and is suitable for high and steep rock slopes, enabling intelligent identification of rock structural surfaces in large-scale mountainous areas. ? 2025, The Authors. All rights reserved.