Probability Evaluation of Rockburst Tendency Considering the Spatial Variation in Rock Mass Properties
LIU Jiazhu,GAO Yongtao,WU Shunchuan,WU Zhongguang,LI Guangquan,LI Zaili
DOI: https://doi.org/10.13374/j.issn2095-9389.2023.04.18.001
2024-01-01
Chinese Journal of Engineering
Abstract:Rockburst disasters pose an increasing threat to the construction safety of deep-buried engineering; thus, rockburst prediction is crucial for ensuring construction safety. However, due to the spatial variation in mechanical properties of rock mass, the actual results of rockburst prediction remain uncertain to some extent. In this study, rockburst tendency and its probability were studied to explore a more suitable evaluation method for rockburst tendency in engineering practice. First, an improved cohesion weakening–friction strengthening model was developed considering the dynamic change of rock dilatancy strength, and the rockburst tendency analysis was combined with the energy index. The point estimation-finite element analysis method was used to analyze rockburst tendency based on the Dahongshan copper mine project buried at a depth greater than 1,000 m. A finite element model was constructed, in which initial cohesion, residual cohesion, residual friction angle, viscous plastic strain critical value, critical value of cohesion plastic strain, and critical value of friction angle plastic strain were used as input variables, and rockburst depth, range, and local energy release value were used as output variables. The specific methods and steps of problem analysis were also elucidated. Furthermore, the probability model of rockburst failure was obtained, and the probability density function and cumulative distribution function were obtained. The probability distribution of the rockburst area was obtained by meshing the failure elements and weight combinations of different scheme results. The research results revealed that the constitutive model and index can better represent the rockburst damage compared with other methods. After considering the variability of rock mass parameters, the depth of rockburst with 95% confidence is consistent with the depth recorded in the field, and the angle range also agrees, which is more accurate than only the fixed value, thus verifying the feasibility and correctness of the uncertainty analysis. The data predicts the unspecified range and local energy release value. Moreover, different statistical indicators conform to different distribution functions. Normal, gamma, and lognormal distributions are optimal for rockburst depth, angle, and local energy release value, respectively. Thus, based on the analysis indication of depth and range, with 80%, 40%, and 20% as the limits, the probability of tendency can be divided into maximum, large, medium, or small, respectively. The probability distribution map of the rockburst area can more intuitively determine the area and probability of rockburst damage. The research results are significant for rockburst support and risk assessment.