论文部分内容阅读
岩爆是高地应力地区影响地下工程施工的主要地质灾害,岩爆预测已成为地下工程的世界性难题之一。地层的岩性条件和地应力的大小是影响岩爆发生的两个最根本因素,对这两个因素的准确判别将直接关系到岩爆预测的成功与否。根据工程的区域地质资料,利用FLAC3D程序建立了该区域的数值计算模型,并结合工程现场实测点的主应力数据,采用径向基函数神经网络,反演了计算区域的初始地应力场。基于现场岩石力学试验和TSP203探测技术,获取了岩爆高风险区域的地层岩性条件。最后结合地应力反演数据和TSP探测结果,对掌子面前面长距离范围内的岩爆强度进行精细预测。工程实际应用表明,该方法可操作性强,岩爆预测结果与实际开挖情况较吻合。
Rockburst is the main geohazard that affects the construction of underground engineering in the high stress area. Rockburst prediction has become one of the worldwide problems of underground engineering. The lithologic conditions of formation and the magnitude of geostress are the two most fundamental factors that affect the occurrence of rock burst. Accurate determination of these two factors will directly affect the success of rock burst prediction. According to the regional geologic data of the project, the numerical calculation model of this area is established by FLAC3D program. Combined with the principal stress data of the measured points in the project site, the radial basis function neural network is used to inverse the initial geostress field of the calculated area. Based on the on-site rock mechanics test and TSP203 detection technology, the stratigraphic lithology conditions of the rockburst high-risk area are obtained. Finally, based on the inversion data of ground stress and the TSP detection results, the rock burst strength in the long distance in front of the face is predicted finely. The practical engineering application shows that the method is feasible and the prediction results of rockburst are in good agreement with the actual excavation.