深部巷道岩爆加权距离判别法模型的分析和应用
来源期刊：中南大学学报(自然科学版)2012年第10期
论文作者：罗磊 曹平
文章页码：3971 - 3975
关键词：岩爆；岩体工程；判别因子；距离判别法；加权距离判别法
Key words：rockburst; rock mass project; discriminant gene; distance discriminant analysis; weighted distance discriminant analysis
摘 要：根据距离判别理论，以巷道最大切向应力和岩石单轴抗压强度的比值、岩石单轴抗压强度和岩石单轴抗拉强度的比值以及弹性能量指数为判别因子，并应用主成分分析法计算权矩阵，建立加权距离判别模型。利用冬瓜山铜矿深部岩体工程实例为学习样本进行训练，通过判别模型的学习功能确定岩爆与其各影响因素之间的复杂关系。用该模型对金川深部岩体工程进行判别，其结果与实际情况基本一致，说明此模型对深部岩体工程岩爆的预测具有较强的实用性。
Abstract: Based on the principle of Mahalanobis’ distance discriminant analysis, the ratio of maximum tangent stress of excavation to rock uniaxial compressive strength，the ratio of rock uniaxial compressive strength to rock uniaxial tensile strength and the index of elasticity energy are used as discriminant gene and the principle component analysis method is adopted to calculate weight matrix to establish weighted distance discriminant model. The deep rock mass engineering of Dongguashan copper mine is taken as the training and testing samples. The complex relations between rockburst and each influencing factor are obtained through the learning function of the discriminant model. Rockburst of the deep rock mass engineering of Jinchuan is used to verify this model, and the result shows that this model has strong practicability to forecast rockburst of the deep rock mass.
罗磊，曹平
(中南大学 资源与安全工程学院，湖南 长沙，410083)
摘 要：根据距离判别理论，以巷道最大切向应力和岩石单轴抗压强度的比值、岩石单轴抗压强度和岩石单轴抗拉强度的比值以及弹性能量指数为判别因子，并应用主成分分析法计算权矩阵，建立加权距离判别模型。利用冬瓜山铜矿深部岩体工程实例为学习样本进行训练，通过判别模型的学习功能确定岩爆与其各影响因素之间的复杂关系。用该模型对金川深部岩体工程进行判别，其结果与实际情况基本一致，说明此模型对深部岩体工程岩爆的预测具有较强的实用性。
关键词：岩爆；岩体工程；判别因子；距离判别法；加权距离判别法
LUO Lei, CAO Ping
(School of Resources and Safety Engineering, Central South University, Changsha 410083, China)
Abstract:Based on the principle of Mahalanobis’ distance discriminant analysis, the ratio of maximum tangent stress of excavation to rock uniaxial compressive strength，the ratio of rock uniaxial compressive strength to rock uniaxial tensile strength and the index of elasticity energy are used as discriminant gene and the principle component analysis method is adopted to calculate weight matrix to establish weighted distance discriminant model. The deep rock mass engineering of Dongguashan copper mine is taken as the training and testing samples. The complex relations between rockburst and each influencing factor are obtained through the learning function of the discriminant model. Rockburst of the deep rock mass engineering of Jinchuan is used to verify this model, and the result shows that this model has strong practicability to forecast rockburst of the deep rock mass.
Key words:rockburst; rock mass project; discriminant gene; distance discriminant analysis; weighted distance discriminant analysis
