基于PCA法和Fisher判别分析法的岩体质量等级分类

岩体质量等级分类在实际的工程中有着很重要的作用。基于主成分分析（PCA）法和与Fisher判别分析法相结合建立岩体质量等级判别模型，选取单轴抗压强度、岩体体积节理数、声波纵波速度、节理面风化变异系数、节理面粗糙度系数和透水性系数6种指标作为岩体质量分级判别的判别因子。以永平铜矿露天矿区工程岩体特征资料中的20个样本为训练样本，10个为待判样本，对该模型进行检验和应用，并与传统的RMR法、Fisher判别分析模型的结果进行比较，相应正确率分别为87%、70%、77%，判断结果表明利用主成分分析法和Fisher判别分析法建立的模型判别能力更高。

Rock mass quality classification based on PCA and Fisher discrimination analysis

Rock mass quality classification plays a very important role in the practical engineering. Based on the combination of principal component analysis(PCA) and Fisher discrimination analysis method, a rock mass quality grade discriminant model is established. Uniaxial compressive strength, rock volumetric joint count, sonic compressional wave velocity, the jointed surface weathering coefficient of variation, the jointed surface roughness coefficient and permeability coefficient are selected as the discrimination factor of the rock mass quality classification. 20 training samples were selected from engineering rock characteristics data in the area of open pit of Yongping Copper Mine and 10 test samples is used to test the model and application. Moreover, comparing with the traditional RMR law and Fisher discriminant analysis, the corresponding correctness rates are 87%, 70%, 77%. The results show that: the combination of principal component analysis and Fisher discrimination analysis method is more accurate than the traditional one.