论文部分内容阅读
提出了在煤田勘探阶段利用主要物理力学性质指标和岩体质量指标进行可采煤层顶底板岩体质量分类的模型.以新疆伊犁州尼勒克县某煤矿为研究背景,选取了天然单轴抗压强度、饱和单轴抗压强度、天然抗拉强度、饱和抗拉强度、天然抗剪强度、饱和抗剪强度、天然密度、含水率、孔隙率、软化系数和RQD值等11项因子作为分类指标,以该煤矿8个可采煤层顶底板的50组岩样样品作为学习样本,首先采用因子分析和Q型聚类分析方法对样品进行综合评价分类,分类结果理想;然后分别计算每一类别样品11项分类指标的平均值并进行对比分析,结果表明,11项指标选取合理;最后采用Fisher判别分析方法对50组样品进行训练,建立了相应的Fisher判别模型,经检验正确率达到98%;将该判别模型应用到乌鲁木齐市某一煤矿对岩样进行分类,其分类结果与神经网络模型、因子和聚类联合分析的分类结果一致,有3个分类结果与规范分类出现偏差,经分析认为新建的判别模型分类结果更为真实可靠.
A model of rock mass classification of top and bottom of coal seam with top and bottom can be selected by using the indexes of main physical and mechanical properties and rockmass quality index in coalfield exploration stage.Using a coal mine in Nilik County, Ili Prefecture, Xinjiang as research background, 11 factors such as compressive strength, uniaxial compressive strength, natural tensile strength, saturated tensile strength, natural shear strength, saturated shear strength, natural density, water content, porosity, softening coefficient and RQD value According to the classification index, 50 rock samples from 8 coal seam roof and floor of the coal mine were taken as learning samples. Firstly, factor analysis and Q-type cluster analysis were used to classify the samples and the classification results were ideal. Then, The average of 11 classification indexes of one sample was compared and analyzed. The results showed that 11 indexes were selected reasonably. Finally, Fisher discriminant analysis was used to train 50 groups of samples, and the corresponding Fisher discriminant model was established. 98%. The discriminant model is applied to a mine in Urumqi to classify the rock samples. The classification results are related to the neural network model, factor and clustering The results of the co-analysis are consistent, there are three categories of results deviate from the normative classification, the analysis that the new discriminant model classification results more real and reliable.