隧道软弱围岩强度特性及支护结构稳定性研究

发布时间：2018-06-20 00:09

本文选题：隧道 + 软弱围岩　；参考：《辽宁工程技术大学》2015年博士论文

【摘要】：软弱岩体的存在给大断面隧道支护结构设计及围岩稳定性计算等带来一系列棘手的问题。由于软弱岩体中节理、裂隙的存在,其力学行为受岩石强度、结构面状况及应力状态等多方面影响,存在节理的硬岩在高应力作用下仍会表现出明显的大变形特性及应变软化特性,岩体的扩容现象突出,岩体变形具有不连续性。所以,对于软弱岩体强度特性及峰后力学行为等问题值得深入研究与讨论。论文针对以上问题,从隧道软弱围岩的变形破坏机制出发,通过室内岩石三轴试验及现场超前地质探测技术相结合的方法,探讨软弱围岩的强度特征,确定围岩结构特性,分析深埋隧道软弱围岩在卸载过程中的应变软化行为及强度损伤演化规律,研究围岩在二次应力作用下的应力场、位移场分布规律；采用收敛-约束法确定围岩及支护结构安全系数,并对支护结构及隧道开挖方法进行优化设计,通过现场监测及数值计算验证优化结构的安全稳定性。论文得出具体结论如下：(1)基于量化GSI围岩评级系统,确定隧道软弱围岩在峰值及峰后残余状态的强度参数,结合Hoek-Brown准则及岩体连续介质应变软化理论,构建深埋软弱围岩峰值及峰后残余强度参数的函数关系式。(2)通过岩石室内三轴试验,确定峰后软化模量与围压关系,构建完整岩石峰后应变软化模型,结合超前地质预报方法,确定深埋隧道软弱围岩力学参数,指导隧道开挖方案优化及支护结构设计。(3)通过数值计算与现场监测数据可知,收敛-约束法适用于大断面软弱隧道围岩支护结构设计,通过FLAC3D有限差分软件构建隧道三维数值模型,得出不同开挖方法的围岩-支护结构相互作用关系,提出各开挖方案的围岩安全系数,并对现有支护方案及开挖方法进行优化设计,通过现场监测数据验证优化设计的可靠性。(4)提出进化神经元算法与BP神经网络算法耦合的GA-BP神经元算法,应用于铺子山隧道施工围岩大变形预测,对隧道围岩变形监测数据进行多步滚动预测,通过实测数据与预测结果相比较,表明该算法具有极高的预测精度,可对类似工程提供借鉴。
[Abstract]:The existence of weak rock mass brings a series of thorny problems to the design of large section tunnel support structure and the calculation of surrounding rock stability. Because of the existence of joints and fissures in weak rock mass, its mechanical behavior is affected by rock strength, structural plane condition and stress state, etc. The jointed hard rock will still exhibit large deformation and strain softening characteristics under the action of high stress. The expansion of rock mass is prominent and the deformation of rock mass is discontinuous. Therefore, the strength characteristics and post-peak mechanical behavior of weak rock mass are worthy of further study and discussion. Aiming at the above problems, starting from the mechanism of deformation and failure of soft surrounding rock in tunnel, this paper discusses the strength characteristics of soft surrounding rock and determines the structural characteristics of surrounding rock through the combination of indoor triaxial test of rock and advance geological exploration technology. The strain softening behavior and strength damage evolution law of soft surrounding rock in deep buried tunnel during unloading are analyzed, and the stress field and displacement field distribution law of surrounding rock under secondary stress are studied. The safety factor of surrounding rock and supporting structure is determined by convergence-constraint method, and the optimum design of supporting structure and tunnel excavation method is carried out. The safety and stability of the optimized structure are verified by field monitoring and numerical calculation. The concrete conclusions are as follows: (1) based on the quantitative GSI surrounding rock rating system, the strength parameters of the peak and post-peak residual state of the soft surrounding rock in tunnel are determined, combined with the Hoek-Brown criterion and the strain softening theory of rock mass continuum. The relationship between post-peak softening modulus and confining pressure is determined through laboratory triaxial tests of rock, and a complete post-peak strain softening model is constructed, combined with advanced geological prediction method. To determine the mechanical parameters of soft surrounding rock of deep buried tunnel, to guide the optimization of excavation scheme and the design of supporting structure. (3) through numerical calculation and field monitoring data, we can know that the convergence-constraint method is suitable for the design of surrounding rock supporting structure of large section weak tunnel. The 3D numerical model of tunnel is constructed by FLAC3D finite difference software, and the interaction relationship between surrounding rock and supporting structure of different excavation methods is obtained, and the safety factor of surrounding rock of each excavation scheme is put forward. The existing support schemes and excavation methods are optimized, and the reliability of the optimization design is verified by field monitoring data. A GA-BP neuronal algorithm coupled with the BP neural network algorithm is proposed, which is coupled with the evolutionary neuron algorithm (ENA) and the BP neural network algorithm (BP neural network algorithm). The method is applied to the prediction of large deformation of surrounding rock in the construction of Pu Zishan Tunnel. The multi-step rolling prediction of the monitoring data of surrounding rock deformation of the tunnel is carried out. The comparison between the measured data and the prediction results shows that the algorithm has a very high prediction accuracy. It can be used for reference for similar projects.
【学位授予单位】：辽宁工程技术大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U451.2

【相似文献】