共查询到20条相似文献，搜索用时 187 毫秒
1.
在砂土液化多指标综合判别法中,采用的判别因子数量一般为5~12,为防止特征冗余,引入核主成分分析(KPCA)对原始样本进行非线性特征提取,同时基于高斯过程分类(GPC)原理,构建了砂土液化预测的KPCA-GPC模型。以唐山地震砂土液化的25个案例为样本,选取平均粒径D50、地下水位dw、标准贯入击数N63.5、砂层埋深ds、地震烈度I、震中距离L、不均匀系数Cu、剪应力与有效上覆应力比τ_d/σ'v共8个指标作为判别因子,对该模型进行验证。研究结果表明:距离判别分析(DDA)、高斯过程分类(GPC)、Seed法的判别准确率分别为83%、83%、67%,而KPCA-DDA和KPCA-GPC的判别准确率均为100%,由此说明消除特征冗余的必要性及KPCA-GPC模型的适用性;同时,与DDA、SVM、BP等确定性判别方法相比,GPC可获得具有概率意义的砂土液化预测结果。 相似文献
2.
《岩土力学》2020,(Z1)
倾斜矿体开采中矿柱的应力状态分析是确保矿体安全开采的重要工作之一。以岩体力学理论为基础,建立了压剪荷载作用下倾斜矿柱的力学模型,推导出表征压剪受载矿柱应力状态的广义Mohr圆方程,分析了矿体倾角、荷载对广义Mohr圆圆心位置及半径的影响,揭示了倾斜矿柱应力演化的3种路径,并证明了3种应力路径分别满足直线方程、圆方程和平面方程。针对矿柱的3种应力路径,提出相应的应力路径演化主控模式。基于Mohr强度理论,分析3种主控模式下的应力路径对矿柱稳定性的影响,阐释倾斜矿柱比垂直矿柱更容易失稳的原因。最后,采用数值模拟方法研究得到矿柱的应力状态与采用广义Mohr圆分析得到的结果一致,且压剪受载矿柱的破坏模式为非对称的沙漏状。研究结果可为倾斜矿柱的稳定性分析提供一种思路。 相似文献
3.
影响砂土液化的因素有很多,建立多指标的液化预测模型非常有必要。目前所有的多指标砂土液化预测模型,均默认选取的判别因子之间相互独立,不存在相关性,可能导致各判别因子之间存在信息叠加而发生误判。以唐山地震砂土液化的25个案例为样本,选取8个影响因素作为砂土液化预测的初始判别指标,首先采用主成分分析（PCA）对各判别指标进行分析,对存在相关性比较高的指标进行了降维处理。基于降维后的4个主成分换算得到新的样本数据,以18个案例为学习样本,建立主成分分析与距离判别分析（DDA）相结合的砂土液化预测模型。利用建立的预测模型对18个案例进行回判,结果全部正确。对其他7个案例的液化情况进行了预测,并与规范法、Seed方法、BP法、DDA法的判别结果进行分析比较,结果表明基于主成分分析与距离判别方法的砂土液化判别模型预测准确率为100%。将模型应用于工程实例,判别结果也与实际情况一致,表明该模型具有良好的预测功能,可在实际工程中应用。 相似文献
4.
依靠岩浆构造环境的地球化学成分认识岩浆形成过程是岩石地球化学中的重要应用。当前利用岩石地球化学成分判别构造环境的工作还不够深入。用4种基于决策树的机器学习方法对来自全球新生代洋岛玄武岩(OIB)、岛弧玄武岩(IAB)及大洋中脊玄武岩(MORB)等镁铁-超镁铁岩中单斜辉石的13种主量元素构成数据集进行了岩浆构造环境判别和主要特征排序。通过对比4种基于决策树的机器学习方法,验证了树类算法对于地球化学成分识别问题的有效性,并总结出4种方法在处理岩浆构造环境判别问题时的优劣:决策树算法判别过程更易于理解,但是其准确率欠佳;boosting算法中的AdaBoost和GBDT对于岩浆构造环境的鉴别准确度较高,但构造过程复杂;bagging集成算法随机森林在权衡性能和模型可理解性时不失为一个良好的选择。此外,还通过4种算法的特征重要性排序得出Cr_2O_3,TFeO,TiO_2,FeO和Al_2O_3是进行岩浆构造环境判别的重要成分。 相似文献
5.
岩石抗压强度是评估岩体工程稳定性的重要力学参数,传统统计回归方法对于岩石抗压强度预测存在一定的局限性.为此,提出了一种利用简单岩石力学参数实现岩石抗压强度智能预测的方法,首先收集了620组含不同类型岩石的三轴试验数据,然后分别采用随机森林(Random Forest,RF)、极限梯度提升树(XGBoost,XGB)和轻量梯度提升机(LightGBM,LGB)3种主流的集成学习算法建立了岩石抗压强度预测模型,使用贝叶斯优化算法在模型训练过程中进行超参数优化,最后利用决定系数(R2)、平均绝对百分比误差(MAPE)和均方根误差(RMSE)对优化后模型的泛化能力进行了综合评估和对比分析.此外,利用LGB模型对输入特征进行重要性分析,以评估不同输入特征对模型泛化性能的影响重要程度.研究结果表明：所建立的3种模型对岩石抗压强度均取得了较好的预测结果,其中LGB模型泛化性能优于另外两种模型(R2=0.978,RMSE=5.58,MAPE=9.70%),且运行耗时相对最少.弹性模量(E)、围压(σ3)和密度(ρ)对模型的泛化性能影响较... 相似文献
6.
利用有限差分程序FLAC建立连续和并排的两类非均质矿柱模型,对矿柱在动力载荷循环作用下的破坏过程进行全时程动力分析。将加卸载响应比(LURR)理论引入矿柱动力稳定性分析中,以施加的动力载荷对矿柱作用的周期性变化作为加载量,以矿柱特征点的位移变化和破坏单元的弹性应变能的演化规律作为响应参数,建立矿柱的加卸载响应比模型,确定矿柱在动力载荷作用期间的加卸载响应比时间序列。计算结果表明,矿柱整体上处于稳定状态和趋于失稳时,其加卸载响应比Y值表现出不同的特征,通过对Y值的分析对矿柱的稳定状态进行评价,预测是否临近峰值承载能力,表明在矿柱动力破裂过程数值模拟分析中引入加卸载响应比是矿柱破坏力学行为研究的一种方便而有效的方法。 相似文献
7.
岩石与矿物的地球化学成分数据具有高维度特征。传统的岩矿地球化学成分研究主要采用二元/三元图解判别法,准确率不高,在数理统计方法上有欠缺。机器学习方法非常适用于对大样本高维度的岩矿成分数据进行数理统计处理。笔者等在介绍机器学习常见算法基本原理的基础上,总结近5年来国内外学者将机器学习方法应用于岩石矿物成分数据研究的实例,包括：(1)根据矿物成分溯源其母岩(源岩)、判别矿床类型,(2)新生代火山岩溯源,(3)判别变质岩原岩,(4)依据岩浆岩成分判别大地构造环境等。已有的研究实例显示,机器学习方法的准确度明显优于传统的低维度判别法。机器学习本质是分析大样本数据的高维度变量之间的相关、归类等多元统计问题。推广机器学习的应用需要建设开放获取(Open Access)的矿物、岩石成分数据库,同时全面实施开放研究(Open Research)的发表策略。 相似文献
8.
为预防地下矿山采空区群系统高应变能的突然释放引发的灾害,通过构建顶板-矿柱三维空间力学模型,利用突变理论和流变力学理论对其稳定性及突变倾向性作定量与定性分析,推导出系统在不同时刻突变并释放能量的数学判据和力学条件,提出系统稳定性的分析算法并验证了其有效性和实用性,在此基础上探讨各影响因子对该系统稳定性的影响。结果表明：随着矿柱流变,系统突变倾向性减小,但顶板整体性逐渐破坏且边界条件依次进入固支、简支和自由边3个阶段,维持系统稳定的矿柱有效承载面积比率临界值 仅在各阶段内连续减小,在顶板简支和自由边的阶段起点处 值突跳增大,3个阶段 值的平均减小速率依次降低;顶板刚度D、上覆岩层荷载 、矿柱面积比率 和空区群尺寸之间的数值关系对系统稳定性起主导作用。该研究结果可为矿山安全开采规划和采空区群系统稳定性评判及调控提供新思路和新方法。 相似文献
9.
基于PCA-GEP算法的边坡稳定性预测 总被引：5，自引：1，他引：4
提出一种基于主成分分析的基因表达式程序设计算法,并将其用于边坡稳定性预测。该算法先采用主成分分析法对样本数据进行预处理,有效地减少预测模型的输入量,消除输入数据间的相关性,再将得到的新样本数据输入基因表达式,构建边坡稳定性的预测模型。利用该预测模型对82个危险圆弧破坏边坡实例中的71个实例进行学习,对另外11个实例进行预测,取得了较好的效果。在保留传统的以误差值作为评判模型优劣标准的同时,引入AIC信息准则法,分别对v-SVR算法和GA-BP网络算法和PCA-GEP算法三种预测模型进行比较分析,结果表明,运用该算法可以获得更优的预测模型,其预测结果比v-SVR算法和GA-BP网络等其他算法得到的结果具有更高的预测精度。工程实例计算表明,该方法是合理、可行的。 相似文献
10.
回采巷道的稳定对煤矿的安全、高效生产至关重要,在生产前应对其评价,以利于采取针对性的措施。为提高回采巷道稳定性预测的准确率,将主成分分析(PCA)与随机森林(RF)算法相结合,对回采巷道稳定性进行预测。根据回采巷道稳定性的影响因素,选取围岩强度、埋深、节理裂隙发育程度、巷道跨度、直接顶与煤层厚度之比和松动圈厚度6个指标作为巷道稳定性的影响因素。采用PCA提取影响因素的主成分,选取贡献率大于80%的3个主成分,代替原有的6个影响因素,将其作为随机森林(RF)的输入变量。将回采巷道稳定划分为4个等级,建立PCA-RF回采巷道稳定性预测模型。研究结果表明:采用PCA-RF模型误判率低,具有较高的预测精度,能够相对有效地对回采巷道的稳定性进行判定。 相似文献
11.
采前煤柱稳定性研究是工作面冲击危险性评估和开采方案设计的关键。以山东某矿深井巨厚砾岩条件工作面开采遗留煤柱为背景,采用案例调研、理论分析、数值模拟和工程实践等方法,对巨厚岩层-煤柱协同变形机制及其煤柱稳定性进行了研究,建立了巨厚岩层-煤柱协同变形的简化力学模型,探讨了引起煤柱变形的主要应力来源和变形形式,推导了在协同变形条件下煤柱的应力-应变关系。以此为基础,综合煤柱煤体应力、围岩稳定性和变形特征等条件,提出了煤柱整体失稳的力学判据。研究结果表明：巨厚岩层-煤柱失稳诱发冲击与煤柱的位置、尺寸和上覆岩层运动或变形关系密切,上覆岩层运动或变形是诱发煤柱失稳的动力因素;巨厚岩层-煤柱的变形主要包括受集中力F压迫的协同挠曲压缩变形和受集中力G作用的重力沉降变形,二者保持内在协调性;巨厚岩层下煤柱整体失稳的工程判据为煤柱煤体平均支承应力p超过其平均极限支承强度Rc（p≥Rc）;评估得到遗留的50 m煤柱具有强冲击危险性,并通过优化开采设计,取得了良好的效果。该研究成果对相似条件煤柱留设及其稳定性分析具有参考意义。 相似文献
12.
Huaizhan Li Guangli Guo Jianfeng Zha Yan He Zhiyong Wang Shuangyu Qin 《Environmental Earth Sciences》2017,76(20):704
We studied an underground coal gasification technique using strip mining-face mining gasifier controlled retraction injection technology (SMFM). This green mining approach offers several advantages over conventional strip mining but the stress distribution and stability of SMFM operations remain largely untested. In particular, hyperbolic coking pillars are used in SMFM compared with rectangular pillars used in traditional strip mining, which can influence the ultimate bearing capacity and stability. We use numerical simulations to investigate the influence of different factors (arch height, pillar height and width, mechanical characteristics), under the coupling effect of high temperature, on the ultimate bearing capacity of hyperbolic pillars. Our results indicate that arch height has a strong influence on pillar stress, while changes in pillar width and height are less significant. A stability evaluation method is proposed and tested on a case study in Inner Mongolia. Our theoretical results have practical significance for the promotion and application of SMFM. 相似文献
13.
矿柱宽度和充填体自立高度是阶段嗣后充填顺利实施的重要影响因素。基于弹性力学平面应变基本假设,建立阶段嗣后胶结充填体矿柱力学模型并进行理论求解。以某铁矿为工程分析实例,采用控制变量法（CVM）研究矿柱不同宽度、高度条件下,水平应力和剪应力的变化规律。研究表明：矿房极限宽度和高度分别为19.8 m和103.2 m。胶结充填体矿柱水平应力随矿柱高度的增高而逐渐增大;剪应力在矿柱中心位置达到最大,且高度越大,剪应力值也越大。在胶结充填体矿柱与非胶结充填体接触侧剪应力趋于定值。矿柱宽度分别为15、18、20 m时,其剪应力分别为243.8、292.6、325.1 kPa。而产生剪应力的主要原因是非胶结充填体受水平应力作用在与胶结充填体矿柱接触面产生滑动摩擦力所致。 相似文献
14.
A relatively novel technique, artificial neural networks (ANN), is used in predicting the stability of crown pillars left over large excavations. Data for the training and verification of the networks were obtained from the literature. Four artificial networks, based on two different architectures, were used. The networks used different numbers of input parameters to predict the stability or failure of crown pillars. Multi‐layer perceptron networks using mine type, dip of orebody, overburden thickness, pillar thickness, pillar length, stope height, backfill height, Rock Mass Rating (RMR) of the host rock and RMR of the orebody showed excellent performance in training and verification. Adding three more variables, namely pillar width, rock density and pillar thickness to width ratio, showed symptoms of over‐learning without degrading performance significantly. Radial basis function networks were capable of predicting crown pillar behaviour on the basis of few input functions. It was shown that mine type, dip and pillar thickness to width ratio can be used for a preliminary estimation of stability. Copyright © 2006 John Wiley & Sons, Ltd. 相似文献
15.
Numerical Investigation of the Dynamic Mechanical State of a Coal Pillar During Longwall Mining Panel Extraction 总被引：2，自引：1，他引：1
Hongwei Wang Yaodong Jiang Yixin Zhao Jie Zhu Shuai Liu 《Rock Mechanics and Rock Engineering》2013,46(5):1211-1221
This study presents a numerical investigation on the dynamic mechanical state of a coal pillar and the assessment of the coal bump risk during extraction using the longwall mining method. The present research indicates that there is an intact core, even when the peak pillar strength has been exceeded under uniaxial compression. This central portion of the coal pillar plays a significant role in its loading capacity. In this study, the intact core of the coal pillar is defined as an elastic core. Based on the geological conditions of a typical longwall panel from the Tangshan coal mine in the City of Tangshan, China, a numerical fast Lagrangian analysis of continua in three dimensions (FLAC3D) model was created to understand the relationship between the volume of the elastic core in a coal pillar and the vertical stress, which is considered to be an important precursor to the development of a coal bump. The numerical results suggest that, the wider the coal pillar, the greater the volume of the elastic core. Therefore, a coal pillar with large width may form a large elastic core as the panel is mined, and the vertical stress is expected to be greater in magnitude. Because of the high stresses and the associated stored elastic energy, the risk of coal bumps in a coal pillar with large width is greater than for a coal pillar with small width. The results of the model also predict that the peak abutment stress occurs near the intersection between the mining face and the roadways at a distance of 7.5 m from the mining face. It is revealed that the bump-prone zones around the longwall panel are within 7–10 m ahead of the mining face and near the edge of the roadway during panel extraction. 相似文献
16.
17.
侧向支承压力分布、资源回收率以及煤柱和巷道的稳定性是大采高综放面区段煤柱宽度留设要兼顾的因素,为了确定大采高综放面区段煤柱宽度,以某矿8103面为工程背景,首先,采用理论计算和现场应力监测等方法确定大采高综放工作面倾向支承压力分布规律,得出应力降低区宽度约为8 m,原岩应力区为巷帮侧28 m外。其次,采用工程类比方法确定大采高综放工作面巷帮外侧煤体严重破裂区宽度约为4 m。最后,采用FLAC3D数值软件分析了下区段工作面回采时窄煤柱（6、8 m）和宽煤柱（28、30 m）的应力场、位移场及塑性区特征,获得不同煤柱宽度时巷道和煤柱力学特征。研究表明：当煤柱宽度6 m和8 m时,在采动支承压力下煤柱几乎无承载能力,且巷道变形量较大;当煤柱宽度28 m和30 m时,在采动支承压力下煤柱中央仍有一定的弹性核,煤柱保持稳定且巷道变形量较小。综合考虑资源回收、巷道稳定性、次生灾害控制等因素,确定大采高综放工作面区段煤柱宽度为28 m。 相似文献
18.
19.
Qing Yu Jinrong Ma Hideki Shimada Takashi Sasaoka 《Geotechnical and Geological Engineering》2014,32(4):821-827
In order to find the relationship between the shaft lining stability and the coal extraction operation, a 3D numerical model of strata layers and shaft lining was established for simulating the influence of coal extraction operation on shaft lining. Certain factors including mining depth, safety pillar width, mining width and mining height were taken as the influence factors in the simulation. The results indicated that the coal extraction could lead to the initiation of the failure in the aquifer and rock layers. As the mining depth increases, the shear strain increment in aquifer becomes small. In this case, the distance between mining panel and aquifer should be larger than 220 m and the safety pillar width should not <70 m. The maximum principal stress in aquifer had a little relation to mining operations. The mining panel width should not exceed 50 m without any support. 相似文献
20.
Naguleswaran Niroshan Ling Yin Nagaratnam Sivakugan Ryan Llewellyn Veenstra 《Geotechnical and Geological Engineering》2018,36(4):2171-2187
This paper is an attempt to relate the microstructure to long-term mechanical properties of the cemented paste backfill produced from a hard rock mine tailing from North Queensland in Australia bound with flyash-based geopolymer (geopolymer), flyash-blended cement (FBC), and general purpose cement (GPC). A relatively high slump (260 mm) paste backfill mix with 74 wt% solids has been used to prepare cylindrical paste backfill samples with a diameter of 50 mm and a height of 100 mm. The uniaxial compressive strength tests were conducted on all samples after curing for 112 days to obtain their strength, failure strain and Young’s modulus. Fractured samples were examined under scanning electron microscope to understand the failure mechanisms at the microstructural scale. The results show that binders significantly affected the mechanical properties of paste backfills (ANOVA, p < 0.05). The paste backfill bound with geopolymer gave the lowest strength and Young’s modulus, while the paste backfills bounded with FBC and GPC showed comparable higher strength and modulus values. This was attributed to the relatively well-packed paste backfills with less cracks and smaller pore sizes in these paste backfills bound with FBC and GPC binders. In particular, needle-shaped particles, which were originally identified in GPC, highly influenced the mechanical property of paste backfills. These results indicate that fly ash can be used to partially replace the cement as a binder for paste backfills to achieve economic and environmental benefits. 相似文献