工程岩体质量分级的数量化理论Ⅱ判别模型及应用

Theory of quantificationⅡ discrimination model and its application to classification of engineering quality of rock mass

为快速准确地对岩体质量进行分级,采用数量化理论Ⅱ建立了工程岩体质量判别模型。案例一中选取岩石单轴抗压强度、岩体声波纵波速度、体积节理数、节理面粗糙度系数、节理面风化变异系数和透水系数作为评价指标,案例二中选取回采巷道巷道埋深、巷道跨度、直接顶与煤层厚度之比、围岩单轴抗压强度、围岩节理裂隙发育程度和围岩松动圈尺度作为评价指标,分别对案例一、案例二中不同工程岩体实测样本进行训练,构建相应的线性判别函数,确定各类岩体样本在6维空间中所对应的样本中心点;用建立的模型对训练样本进行判别并对未知样本岩体质量等级进行了预测。案例一中判别正确率为100%,且10个预测样本中只有1个发生误判,案例二中判别正确率为93.3%。研究结果表明:该模型判别过程简单,具有较高的判别正确率,可以作为一种行之有效的岩体质量分级方法在岩体工程中推广应用。

In order to classify th el of engineering quality of rock e quality of rock mass rapidly and accurately, a discrimination mod- mass was established based on xes including uniaxial compressive strength theory of quantification Ⅱ. Six inde- , P-wave velocity in rock, volumetric joint count, rough- ness coefficients of joints, weathering coefficients and permeability coefficient of joints were selected as the evaluation indictors in case No. 1. Depth of the mining roadway, span of the mining roadway, ratio of the immediate roof seam to the coal seam thickness, intensity of rock, rock joint development degree and the loose ring size were selected as the evaluation indicators accordingly in case NO. 2. Linear discrimination functions were established and central points of different kinds of sam- ples in the six-dimensional space were obtained by training different sets of practical engineering samples. Then the model was used to test the training samples and predict the classification of the unknown samples. The ratio of right-predicted was 100% and only one of the unknown samples was predicted wrongly in case No. 1. The ratio of right-predicted in case No. 2 was 93.3%. The resultssuggest that this discrimination model, discriminating the samples easily and accurately, can be widely used in practical engineering.