Discrete element method and support vector machine applied to the analysis of steel mesh pinned by rockbolts. (January 2020)

• Record Type: Journal Article
• Authors: Xu, Cong; Tannant, Dwayne D.; Zheng, Wenbo; Liu, Kaiyun
• Is Part Of: International journal of rock mechanics and mining sciences. Volume 125(2020)
• Publication Date: 2020-01
• Subjects: Bolted steel mesh -- Rock block -- Discrete element method -- Support vector machine; Rock mechanics -- Periodicals; Soil mechanics -- Periodicals; Mining engineering -- Periodicals; Roches, Mécanique des -- Périodiques; Sols, Mécanique des -- Périodiques; Technique minière -- Périodiques; 624.151305
• DOI: 10.1016/j.ijrmms.2019.104163
• Languages: English
• ISSNs: 1365-1609

Abstract

Abstract: The current designs of steel mesh pinned by bolts on steep rock slopes do not adequately consider the deformation of the mesh. This paper discusses the development of discrete element models of wire mesh and a moving rock block to simulate field conditions where an unstable rock located between the bolts slides along a joint to load the mesh. A series of parametric analyses were performed using 900 simulations to analyze the influences of various factors on the mesh response. A high tensile-strength steel wire mesh with 3 mm diameter wires was selected for the analysis. The mesh wires are more likely to rupture if the rock volume is larger than 1.5 m 3 and the sliding angle is over 40°. For the scenarios that were simulated, the steel mesh typically experienced a bulge of 0.4–0.8 m when loaded by rock volumes up to 3 m 3 . A larger mesh bulge occurs when the loose rock originates close to the upper bolt in the surrounding pattern of bolts. The mesh bulge increases dramatically as the bolt spacing increases. The simulation results were used to create mesh performance prediction models via a support vector machine (SVM) approach. The design tool considers a mesh bulge of 0.5 m and wire rupture as design criteria. The simulation results were used to train and test SVM models to predict the mesh response for a wide range of influence factors.