基于实测数据的长隧道地应力分布影响因素研究

地应力分布特征是开展长隧道动态优化设计与施工的主要影响因素，亦是判断与预防岩爆、大变形等隧道灾害的重要依据。为了研究某长隧道区域地应力分布的影响规律，以17个现场水压致裂钻孔、164个测点的地应力测试数据为基础，采用高程、岩性、结构面、沟谷地形分区方法分析了长隧道区域的地应力分布特征。通过对比分析不同分区下长隧道围岩应力拟合值与实测值的差异，认为根据地质特征对长隧道区域钻孔测试数据进行合理分区是准确拟合围岩地应力分布特征的关键。研究结果表明：在岩性分界面两侧地应力呈现不同分布规律，沟谷地形对应力分布规律影响显著，而高程和小节理对应力分布的影响较小；岩性分区的线性拟合方法适用于非沟谷地形区域的地应力分布规律分析，相较于按整体或单因素的分区方法得到的地应力分布，采用线性和曲线相结合的沟谷分区方法推算的隧道围岩应力分布具有较高的精度；由于沟谷地形在一定埋深区域存在高应力区，即便采用线性和曲线拟合方法也难以准确模拟其真实的应力分布特征，需进一步开展针对性的研究。

Influencing Factors of In-situ Stress Distribution in Long Tunnel Based on Measured Data

Distribution of in-situ stress is the main factor influencing the dynamic optimization of design and construction of long tunnels, and is also an important basis for the judgment and prevention of tunneling disasters such as rock bursts and large deformations. To study the influencing factors of in-situ stress distribution in a long tunnel area, the in-situ stress test data of 164 measuring points in 17 hydraulic fracturing boreholes are statistics via partition methods considering buried depth, lithology, rock mass structure and valley topography, and the in-situ stress distribution characteristics in a long tunnel area are analyzed. By comparing the difference of in-situ stress between the fitted values and measured values in different partitions, it is found that reasonable partition of borehole test data in a long tunnel area according to geological characteristics is the key to the accurate fitting of the in-situ stress surrounding the tunnel. Results show that: Valley topography is an important influence on stress distribution and distribution of in-situ stress on both sides of lithologic interfaces shows different regularity, while height and small joints have little effect on in-situ stress distribution. The linear fitting method considering lithology is suitable for the analysis of in-situ stress distribution in non-valley areas. Compared with the fitted results considering the integral area or a single factor, those fitted via a combination of linear or curve functions have higher accuracy in valley topography. Since there is a high-stress area in a certain buried depth in the valley topography, it is difficult to accurately simulate the in-situ stress distribution characteristics even by using the linear and curve fitting methods, so further targeted research is needed.