中文摘要：

为快速有效地预测隧洞围岩的类别,提高地下工程的稳定性和安全性,应用因子分析与Fisher判别分析理论,选取岩石质量指标、完整性指标、饱和单轴抗压强度、纵波波速、弹性抗力系数和结构面摩擦因数等6个指标作为Fisher判别分析的判别因子。建立基于因子分析的隧洞围岩分类的Fisher预测模型。将现场勘测的30组隧洞围岩数据作为学习样本进行训练。利用回代估计法对模型效果进行检验,正确率为96.7%。将建立的判别模型应用于工程实例,以6组工程数据作为预测样本,进行隧洞围岩的分类预测,并与神经网络方法和Bayes方法进行对比。结果表明：因子分析可以有效提取围岩分类指标,去除冗余影响因素,基于因子分析的Fisher判别模型可有效地预测隧洞围岩的类别,所得预测结果的正确率为100%。

英文摘要：

In order to predict the tunnel surrounding rock category quickly and effectively and to enhance the stability and safety of underground engineering, applying the theory of factor analysis and Fisher discriminant analysis, and selecting rock quality, integrity, saturated uniaxial compressive strength, longitudinal wave velocity, elastic resistance coefficient and structure surface friction factor as the discriminant factor of Fisher＇s discriminant analysis, a Fisher prediction model for tunnel surrounding rock category based on factor analysis is built. Thirty groups of tunnel surrounding rock data in site survey are used as learning samples for training. The resubstitution predicting method is used to test the model, and the accuracy is 96. 7 %. The established discriminant model is applied to engineering instance, 6 sets of engineering data are taken as test samples to forecast the classification of tunnel surrounding rock, and the outcome is compared with those of neural network method and the Bayes method. The result shows that the factor analysis can effectively extract the classification indexes of surrounding rock and remove redundant influencing factors. The Fisher＇s discriminant model based on factor analysis can effectively predict the tunnel surrounding rock category, and its prediction accuracy is 100%.