基于钻进特征参数的隧道掌子面围岩级别智能识别方法

Intelligent Identification Method of Surrounding Rock Grades of Tunnel Face Based on Drilling Parameters

针对钻爆法隧道传统围岩分级方法操作复杂、受主观因素影响大,在评判复杂岩体的质量等级时存在一定局限性等问题,通过优选核密度估计法拟合钻进参数的分布,提出基于朴素贝叶斯分类算法的隧道围岩级别智能识别方法,采用交叉验证法提升分类模型性能,并依托文献数据开展应用验证。结果表明:基于核密度估计的朴素贝叶斯分类算法可以利用钻进参数的实际情况确定掌子面围岩质量等级,在测试集样本中的分类准确率达到94.0%;基于交叉验证法的分类模型性能提升方法可以充分利用钻进参数-围岩等级数据集的信息,有效提升分类算法模型的性能,在299组测试样本中分类准确率高达98.7%。

To address the complexities and subjectivity in traditional rock mass classification methods for drill-andblast tunnels,which have limitations when assessing complex rock masses,this study proposes an intelligent method for identifying tunnel surrounding rock grades.The method involves fitting the distribution of drilling parameters using the kernel density estimation method and employing a Naive Bayes classification algorithm for rock classification.The performance of the classification model was enhanced through cross-validation,and the method was validated using literature data.The results demonstrate that the Naive Bayes classification algorithm based on kernel density estimation can accurately classify the quality of the tunnel face surrounding rock using drilling parameters,achieving a classification accuracy of 94.0%in the test set.Furthermore,the cross-validation method improved the model′s performance,reaching a classification accuracy of 98.7%on a test set of 299 samples.