基于实测钻凿参数的类岩石材料强度识别研究  

Research on strength identification of rock-like materials based on measured drilling parameters

作　　者：王喆 仇安兵 龚敏[1] 吴昊骏 胡广风 王思杰 周世均 WANG Zhe;QIU Anbing;GONG Min;WU Haojun;HU Guangfeng;WANG Sijie;ZHOU Shijun(School of Civil and Resource Engineering,University of Science and Technology Beijing,Beijing 100083,China;Chongqing Zhonghuan Construction Co.,Ltd.,Chongqing 401120,China)

机构地区：[1]北京科技大学土木与资源工程学院,北京100083 [2]重庆中环建设有限公司,重庆401120

出　　处：《矿业安全与环保》2023年第4期55-62,67,共9页Mining Safety & Environmental Protection

摘　　要：为建立凿岩机钻凿参数与岩石强度间的关系,进行岩石强度识别研究,搭建了台车钻凿参数自动采集系统,选用C30、C40、C50共3种强度混凝土模拟同等强度岩石,动态采集不同钻凿参数,并基于支持向量机(SVM)算法构建4种SVM分类模型,对钻凿数据进行训练学习并运用优化算法修正核函数系数,根据分类准确率及评价指标完成模型优选。研究表明,采用多项式核与高斯核函数建立的SVM模型识别准确率达90%,可以有效识别类岩石材料强度。In order to establish the relationship between drilling parameters of rock drill and rock strength,and carry out research on rock strength identification,an automatic collection system of drilling parameters by truck was built.In the test,three kinds of strength concrete,C30,C40,and C50,were selected to simulate the rock of the same strength,and different drilling parameters were dynamically collected.Four SVM classification models were constructed based on the Support Vector Machine(SVM) algorithm.The drilling data was trained and learned,and the kernel function coefficient was modified by optimization algorithm.The model was optimized according to classification accuracy and evaluation indicators.The results show that the accuracy of SVM model based on polynomial kernel and Gaussian kernel function can reach 90%,which can effectively identify the strength of rock-like materials.

关 键 词：岩石强度 支持向量机 随钻识别 钻凿参数采集 凿岩机 

分 类 号：TD231[矿业工程—矿井建设]