首页|基于微震信息的岩爆体积分级与判别方法研究

基于微震信息的岩爆体积分级与判别方法研究

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
深埋地下工程岩爆灾害的危害性与其爆坑规模直接相关,为进一步提高岩爆灾害的精细化表征及预测水平,开展基于微震信息的岩爆爆坑体积分级及判别方法研究.首先,通过对收集到的111个来源于锦屏二级水电站深埋隧洞群工程的岩爆案例进行统计分析可知,岩爆孕育过程中的累积微震事件数、累积微震释放能、累积微震视体积、微震事件率、微震能量释放率、微震视体积率这6个指标与岩爆爆坑体积之间具有较高相关性,即岩爆爆坑的体积分布与微震参数取值之间呈现出较为明显的从低到高的层次差异性.其次,利用层次聚类分析手段,从工程实用性与可预测性的角度构建一种岩爆体积分级划分方案,以锦屏隧洞工程为例,将岩爆体积等级划分为五级,并确定相应等级的体积阈值.最后,基于改进的分类回归树(classification and regression tree,简称为CART)算法,构建了用于确定不同岩爆体积等级下各微震参数判别阈值的决策树,形成岩爆体积等级的6个单微震参数判据;进一步,提出一种基于多微震参数的岩爆体积等级综合判别的蛛网图方法,并通过案例反分析确定相应的判别准则,利用该方法可快速实现洞室开挖过程中岩爆潜在规模等级的判别.对收集所得岩爆案例进行回溯验证,结果显示,岩爆回判准确率总体达到85.2％,表明该方法具有较高的准确率与适用性.该研究能够为类似深埋地下工程岩爆灾害精细化预测水平的提高提供一种新的有效途径.
Research on rockburst volume classification and discriminant method based on microseismic information
The hazard of a rockburst event is directly correlated with the scale of rock mass ejection in deep underground engineering.In order to further enhance the fine characterization and prediction level of rockburst hazards in deep underground engineering,a study was conducted on the classification and discrimination method of rockburst pit volume based on microseismic information.Firstly,a statistical analysis of one hundred and eleven rockburst cases from the deep tunnels of Jinping Ⅱ hydropower station project was performed.It was found that six indicators,namely the cumulative number of microseismic events,cumulative microseismic energy release,cumulative microseismic volume,microseismic event rate,microseismic energy release rate and microseismic volume rate,showed a high correlation with the volume of rockburst pit.In other words,there was a significant hierarchical difference from low to high between the distribution of rockburst volumes and the values of microseismic parameters.Secondly,a hierarchical clustering analysis was employed to establish a classification scheme for rockburst volumes,taking into consideration engineering practicality and predictability.Taking the Jinping tunnel project as an example,the rockburst volume was divided into five levels,and the corresponding volume thresholds for each level were then determined.Finally,a decision tree based on the improved classification and regression tree(CART)algorithm was constructed to determine discrimination thresholds for various microseismic parameters under different rockburst volume levels,and six individual microseismic parameter criteria for rockburst volume classification were therefore developed.Furthermore,a spider web diagram method based on multiple microseismic parameters was developed for comprehensive discrimination of rockburst classification,and corresponding discrimination criteria were determined through case analysis.This method enables people to rapidly discriminate the potential level of rockburst volume during tunnel excavation.The results from retrospective verification of the collected rockburst cases showed an overall accuracy rate of 85.2％for rockburst volume discrimination,demonstrating a high accuracy and applicability of the proposed method.This research provides a new and effective approach to improve the fine prediction level of rockburst hazards in similar deep underground engineering projects.

rock mechanicsrockburst volumemicroseismic informationdeep tunnelclustering analysisrockburst classificationcomprehensive discrimination

刘国锋、李胜峰、丰光亮、陈炳瑞、许江波、杜程浩、陈学琦

展开 >

长安大学公路学院,陕西西安 710064

中国科学院武汉岩土力学研究所岩土力学与工程国家重点实验室,湖北武汉 430071

岩石力学 岩爆体积 微震信息 深埋隧洞 聚类分析 岩爆分级 综合判别

国家自然科学基金中央高校基本科研业务费专项陕西省自然科学基础研究计划

522091203001022132032022JM-191

2024

10.13722/j.cnki.jrme.2023.0598

岩石力学与工程学报
中国岩石力学与工程学会

岩石力学与工程学报

CSTPCD北大核心
影响因子：2.589
ISSN：1000-6915
年,卷(期)：2024.43(3)
  • 38