Shaft stability has long been a critical concern in rock mass engineering,particularly as it pertains to underground mines.In the Hongbu mining area of the Xincheng gold mine,the 3# shaft serves as the primary hoisting shaft,and its stability is directly correlated with the mine’s operational benefits.With the depletion of shallow resources,mining activities at Hongbu are progressively transitioning to deeper levels,necessitating comprehensive research on the impact of deep mining on the stability of the 3# shaft.To facilitate dynamic evaluation of the shaft's stability and advance risk prediction,several measures have been undertaken.Initially,a deep hole inclinometer was employed to monitor the deformation of the shaft's surrounding rock.Subsequently,an advanced prediction algorithm for surrounding rock deformation wasdeveloped using the Long Short-Term Memory (LSTM) algorithm,alongside a four-color early warning index system for shaft stability based on tangent angle and cumulative deformation.Finally,utilizing Dempster-Shafer (D-S) theory,a dynamic evaluation and risk early-warning method for shaft stability was established,integrating multi-source early-warning index fusion.The findings indicate that Shaft #3 is presently in a stable condition;however,there is a notable inconsistency in deformation at approximately 160 meters,warranting further investigation.The rock mass deformation algorithm,utilizing Long Short-Term Memory (LSTM),successfully achieves advanced predictions of rock mass deformation with an accuracy rate of 99%.The evaluation of shaft stability and the advanced prediction method,grounded in Dempster-Shafer (D-S) theory,reveal that the risk value of the shaft remains below 20% throughout the monitoring period,signifying its current safe state.Furthermore,projections suggest that the shaft will continue to exhibit low risk over the next 30 days.This research provides both theoretical and empirical support for the safe and efficient operation of mining activities and introduces a novel approach to the monitoring,early warning,and risk prediction of shaft stability.
竖井作为矿山开采的咽喉工程,尽管其工程量只占全矿井巷工程量的10%左右,但施工周期往往占矿山建设总工期的40%以上(刘溪鸽,2016;Wang et al.,2022)。作为矿山生产系统的核心载体,竖井围岩的稳定性直接影响出矿、排碴、通风和排水等全流程作业的安全运行。因此,开展竖井围岩力学响应监测并建立动态稳定性评价体系,对保障矿山安全生产具有重要意义。
近年来,竖井稳定性研究已成为岩体工程领域的热点方向(杨官涛等,2009;汪小平等,2021)。现有研究主要依托工程类比、相似模拟、数值仿真及现场监测等方法,重点关注竖井变形影响因素、开采和爆破对竖井的作用机制、断层部位井壁脆弱性评估以及支护优化策略等(赵海军等,2012;刘溪鸽,2016;张洪训等,2017;过江等,2022;罗文俊等,2022)。此外,井壁围岩力学特性和支护优化也是研究的关键方向,约束—收敛法的提出与应用为竖井支护提供了新的理论依据(Oreste,2009;Sandrone et al.,2010)。孙闯等(2013)基于Hoek-Brown准则(Hoek et al.,1997)构建了围岩特征曲线,并结合“高阻让压”理念优化二次支护时机,显著提升了竖井的稳定性。同时,采用解析法并结合数值模拟进一步深化了竖井开挖力学模型的研究(殷有泉等,2014;陈勋等,2020;郝勇浙等,2022),为竖井长期稳定性设计提供了理论支撑。尽管上述研究取得了重要进展,但仍存在以下不足:(1)数据实时性不足,现有研究多依赖数值模拟与理论分析,缺乏围岩变形的实时监测数据支撑,难以动态反映围岩真实力学响应;(2)预测精度受限,传统方法(如数值模拟和工程类比法)在处理复杂非线性变形规律时存在局限性,预测结果与实际变形存在偏差;(3)预警体系不完善,当前研究多聚焦单一指标评价,缺乏多源数据融合的动态预警机制,难以全面评估竖井稳定性风险状态。
在空间特征分析中,通过建立深度—累计位移关联曲线,可有效识别潜在滑移面位置。基于测斜仪监测数据研究(靳晓光等,2001;孙志彬等,2010;Chen et al.,2024),划分出5类典型变形模式:D型、V型、B型、r型和震荡型。其中,D型、B型和r型表明岩体出现了非连续变形,很有可能产生了破裂面或滑移面,岩体处于加速或等速蠕变阶段,具有潜在的风险;当岩体处于V型或震荡型,表明尚未产生破裂面,处于初始蠕变或等速蠕变阶段。该分类体系为围岩稳定性评估提供了量化判据。
当α<45°时,岩体处于减速蠕变阶段,不存在灾变的可能;当α=45°时,岩体进入等速蠕变阶段;当α>45°时,岩体进入加速蠕变阶段,存在发生灾变的可能性。目前该指标已被国内外学者推广应用(Crosta et al.,2003;郭延辉等,2024;许强等,2025),经过大量工程验证,证明了该指标的有效性,并给出了该指标对应的四级预警指标体系,如表2所示。
值得注意的是,当前等速蠕变速率的计算依赖于历史的监测数据,由于目前蠕变曲线并不完善,提取的等速蠕变速率可能存在一定的误差,进而对最终的风险评价结果造成影响(郭延辉等,2024)。针对此问题,本研究一方面通过定期观察监测数据,动态更新等速蠕变速率,以此不断提高等速蠕变速率的精度,另一方面基于现场工程经验和历史案例分析算法(Zhang et al.,2022),确定了累计变形量和变形速率分级预警评价阈值,形成更为综合全面的竖井稳定性评价预警指标体系,如表2所示。在此基础上,采用D-S证据理论,进行多源指标的融合,以解决上述问题,提高风险评价精度。
ChenH, WuH G,2024.Method for determining the position of landslide slip-surface with a typical inclinometric curves[J].Journal of Mountain Science,21(2):413-432.
[2]
ChenXun, YinShenghua, ChenShunman,et al,2020.Influence of deep mining under multilayer goaf on the stability of hanging shaft[J].Mining Research and Development,40(9):27-32.
[3]
CrostaG B, AgliardiF,2003.Failure forecast for large rock slides by surface displacement measurements[J].Canadian Geotechnical Journal,40(1):176-191.
[4]
GuoJiang, JiangNiming, ChengXin,2022.Numerical simulation study on shaft construction process of water-rich moraine layer[J].Gold Science and Technology,30(5):733-742.
[5]
GuoYanhui, HuoYuan, MaoXiaojuan,et al,2024.Research on early warning criterion of landslide critical sliding tangential angle based on improved Nishihara model[J].Journal of Natural Disasters,33(5):96-108.
[6]
HaoYongzhe, ZhangFei, WangHao,2022.Analysis of the influence of multiple goafs on shaft stability based on 3DMine-Rhino-FLAC3D [J].China Mining Magazine,31(2):65-71.
[7]
HoekE, BrownE T,1997.Practical estimates of rock mass strength[J].International Journal of Rock Mechanics and Mining Sciences,34(8):1165-1186.
[8]
JinXiaoguang, WangLansheng, LiXiaohong,2001.Determination and advanced forcasting on slide plane position for landslide[J].The Chinese Journal of Geological Hazard and Control,(1):14-16.
[9]
LiuXige,2016.Numerical Simulation on Effect of Mining of Ⅺ# Ore Body on Stability of Main Shaft at Xincheng Gold Mine[D].Shenyang:Northeastern University.
[10]
LuoWenjun, GuanKai, ZhuWancheng,et al,2022.Influence of blasting vibration near main shaft on its stability in Xincheng gold mine[J].China Mining Magazine,31(5):121-127.
[11]
OresteP,2009.The convergence-confinement method:Roles and limits in modern geomechanical tunnel design[J].American Journal of Applied Sciences,6:757-771.
[12]
SandroneF, LabiouseV,2010.Analysis of the evolution of road tunnels equilibrium conditions with a convergence-confinement approach[J].Rock mechanics and rock engineering,43:201-218.
[13]
SunChuang, ZhangXiangdong, ZhangJianjun,2013.Stability analysis of vertical shaft surrounding rock and supporting system in deep fault fracture[J].Journal of China Coal Society,38(4):587-594.
[14]
SunZhibin, YangXiaoli,2010.Sliding characteristics analysis of land slide based on the deep displacement[J]. Journal of Changsha University of Science and Technology(Natural Science),7(2):43-47.
[15]
WangX, LiuZ, FanY Y,et al,2022.Destruction characteristics and control countermeasure of shaft surrounding rock mass in complex geological environment[J].Sustainability,14:13329.
[16]
WangXiaoping, KouYongyuan, GuoYunlin,et al,2021.Study on the stability of 14 row return air shaft in No.1 ore body hanging wall lean ore of Jinchuan No.2 mining area[J].Me-tal Mine,50(6):165-171.
[17]
XuQiang, ChenGuoqing, WeiTao,et al,2025.Dynamic charac-teristics of stress,sliding force and deformation during the evolution of sudden failure landslide[J/OL].Chinese Journal of Rock Mechanics and Engineering,1-14[2025-01-15].
[18]
XuQiang, TangMinggao, XuKaixiang,et al,2008.Research on space-time evolution laws and early warning-prediction of landslides[J].Chinese Journal of Rock Mechanics and Engineering,(6):1104-1112.
[19]
YangGuantao, LiXibing, LiuXiling,2009.Minimum safety factor method for stability analysis of vertical shaft surrounding rockmass and supporting system[J].Journal of China Coal Society,34(2):175-179.
[20]
YinYouquan, LiPing’en, DiYuan,2014.Instability of shaft excavation and mechanism of rockburst[J].Chinese Journal of Theoretical and Applied Mechanics,46(3):398-408.
[21]
ZhangHongxun, LiuXige, GuanKai,et al,2017.3D delineation of the shaft safety pillar of Xincheng gold mine and its numerical excavation[J].Metal Mine,46(3):19-24.
[22]
ZhangP, DengW, LiuF,et al,2022.Establishment of landslide early-warning indicator using the combination of numerical simulations and case matching method in Wushan open-pit mine[J].Frontiers in Earth Science,10:960831.
[23]
ZhaoBaobing,2024.Prediction of rock pressure in roadway working face based on optimized LSTM network[J].Coal Technology,43(2):54-56.
[24]
ZhaoHaijun, MaFengshan, XuJiamo,et al,2012.Shaft deformation and failure due to rock mass movement induced by underground backfill mining of a metal mine[J].Chinese Journal of Geotechnical Engineering,34(2):340-348.