P波入射下山体−隔震层−隧道的整体响应分析

黄磊, 刘翔, 刘中宪, 段更月, 蒋坪临, 艾天淳

黄磊,刘翔,刘中宪,段更月,蒋坪临,艾天淳. 2024. P波入射下山体−隔震层−隧道的整体响应分析. 地震学报,46(3):526−538. DOI: 10.11939/jass.20230093
引用本文: 黄磊,刘翔,刘中宪,段更月,蒋坪临,艾天淳. 2024. P波入射下山体−隔震层−隧道的整体响应分析. 地震学报,46(3):526−538. DOI: 10.11939/jass.20230093
Huang L,Liu X,Liu Z X,Duan G Y,Jiang P L,Ai T C. 2024. Analysis of the overall response of mountain-isolation layer-tunnel under P-wave incidence. Acta Seismologica Sinica46(3):526−538. DOI: 10.11939/jass.20230093
Citation: Huang L,Liu X,Liu Z X,Duan G Y,Jiang P L,Ai T C. 2024. Analysis of the overall response of mountain-isolation layer-tunnel under P-wave incidence. Acta Seismologica Sinica46(3):526−538. DOI: 10.11939/jass.20230093

P波入射下山体−隔震层−隧道的整体响应分析

基金项目: 国家自然科学基金(52108466,52278516)和天津市教委自然科学基金(2020KJ039)联合资助
详细信息
    作者简介:

    黄磊,博士,讲师,主要从事地震工程波动研究,e-mail:huanglei1289@163.com

    通讯作者:

    刘中宪,博士,教授,主要从事地震工程和工程波动领域的研究,e-mail:zhongxian1212@163.com

  • 中图分类号: 315.31

Analysis of the overall response of mountain-isolation layer-tunnel under P-wave incidence

  • 摘要:

    采用高精度间接边界元方法(IBEM)开展了山体−隔震层−隧道的整体响应分析,研究了P波入射下柔性隔震层对高斯形山体内双线隧道地震动响应的减弱作用。在隔震层与围岩之间考虑了碎土的错动滑移,以模拟不完美交界面,并系统讨论了隔震层弹性模量、厚度、P波入射频率等因素对隧道地震动响应的影响。数值模拟结果表明:适当增加隔震层厚度可以显著降低隧道应力,有效提升隔震效果,应力降低幅度约达50%以上,且使得结构受力更为均匀;随着隔震层材料的弹性模量减小,隧道动应力数值明显减小,有效减少了其应力放大区的面积,从而有效避免因地震而导致的衬砌裂缝。

    Abstract:

    As an important transportation facility, the safety and stability of mountain tunnels are of great concern when natural disasters such as earthquakes occur. Seismic wave reflection and coherence effect will occur in the mountain body, and the mountain body, seismic isolation layer, and tunnel as “scattering body” and “secondary source” can change the spatial distribution of ground shaking distribution and values of seismic ground motion. In order to study the influence of isolation measures on seismic wave scattering in double line tunnels within mountains, we uses high-precision indirect boundary element method (IBEM) to analyze the overall response of mountains, isolation layers, and tunnels under P-wave incidence. The isolation effect of flexible isolation layers on double line tunnels crossing Gaussian-shaped mountains.

    Firstly, a comprehensive computational model was developed. A two-lane lined tunnel is traversed within the Gaussian-shaped mountain, and a seismic isolation layer is set between the tunnel and the mountain site. It is assumed that the mountain site, the tunnel and the seismic isolation layer are all linear isotropic media. In this article, there is not complete consolidation between the tunnel and the surrounding rock, but rather dislocation slip. Therefore, a series of virtual linear springs and dampers are used to connect the isolation layer and the surrounding rock, and a viscoelastic boundary is set to simulate this imperfect boundary.

    Secondly, wave field analysis was conducted based on elastic wave theory. The formulas for the free field and scattering field are provided in the article, and an equilibrium equation is established based on the displacement and stress continuity conditions at the interface of each computational domain. Solving this equation can obtain the virtual wave source density. Multiplying the concentration of the virtual wave source by the corresponding Green’s function can obtain the scattering field of observation points in each domain. The superposition of scattering field and free field yields the full wave field.

    Thirdly, the correctness of the results is also verified. Due to the lack of an accurate analytical solution for mountainous tunnels under P-wave incidence in current research, we degenerates the model into a half space tunnel model and compares it with published results, with the same setting of the calculation parameters. It can be found that the results of this paper are in good agreement with those of published paper, thus verifying the accuracy of this method.

    Finally, the effects of the modulus of elasticity, thickness, and incident frequency of the seismic isolation layer on the seismic ground motion response of the tunnels were discussed in detail, and the displacements and stresses of the tunnel inside the mountain and the displacements on the mountain surface are obtained. The research results can provide some reference for seismic isolation design and construction of double-line tunnel in mountain site. The following main conclusions are obtained:

    1) IBEM can accurately solve the seismic dynamic response of lined tunnels in mountain, including the amplification effect of seismic ground motion in mountain ranges and the stress concentration effect of the lining, etc. Setting up seismic isolation and damping measures can effectively change the distribution of the stress and give full play to the load-bearing capacity of the surrounding rock, thus playing a role in protecting the tunnel lining.

    2) As the modulus of elasticity of the isolation layer material decreases, there is a significant reduction in the values of dynamic tunnel stresses. The structural response of the flexible seismic isolation layer for suppressing high-frequency waves is more obvious. When high-frequency waves are incident, an isolation layer with elastic modulus of 12 MPa can reduce the tunnel peak stress to 25.4% of which without an isolation layer. At the same time, when the modulus of elasticity is small, it can effectively reduce the area of its stress amplification zone, thus effectively avoiding lining cracks caused by seismic action.

    3) The seismic isolation layer can make the lining circumferential stress distribution tend to be uniform, and with the increase of the thickness of the vibration isolation layer, the tunnel lining dynamic stress gets smaller, and the reduction can be about 50% or more.

    4) Considering the efficiency of seismic isolation as well as the cost of the project, among the several sets of parameters studied in this paper, the combined effect of a seismic isolation layer with a modulus of elasticity of 12 MPa and a thickness of 20 cm is superior.

    5) Considering the existence of broken soil between the tunnel and the surrounding rock, a staggered slip boundary is introduced on the boundary. The staggered slip boundary model proposed in this paper is still simplified and does not take into account the nonlinear effects such as elastic-plasticity and large deformation.

  • 相关研究表明,隧道在地震中会遭受严重的破坏(禹海涛等,2022),特别是位于山岭地带的隧道,如汶川地震中的烧火坪隧道、龙洞子隧道等均发生了衬砌开裂和拱顶坍塌的情况(李天斌,2008),极大地降低了救援的效率。因此,在山岭隧道的设计和建设过程中应充分考虑地震因素,并采取一系列隔减震措施以减轻地震对隧道的破坏,这对于确保隧道结构稳定尤为关键。

    目前,隧道工程多采用刚性抗震措施,主要通过改变围岩力学参数的施作锚杆(Bobet,2009)或对围岩体注浆加固(Xu et al,2021),提高结构的自身性能的采用钢纤维混凝土(Meng et al,2016)来减轻地震作用对隧道的影响。然而,这些措施的实施可能会导致隧道结构的局部应力增加,从而引起裂纹甚至坍塌。越来越多的学者开始关注和研究柔性隔减震技术。当前主流技术是在结构与围岩之间设置隔震层(Su et al,2019Ma et al,2020),其作用机理是在结构与围岩之间设置低刚度的耗能装置,使结构柔度增加,自振周期增大。在地震作用下,耗能装置能吸收或隔离大量的地震能量,从而减小地震对隧道的损伤,使隧道结构整体更具有韧性和抗震能力。Jiang等(2018)和Zhou等(2021)通过振动台试验,证明了隔震层可使隧道衬砌结构主要部位的峰值应力减小,并明显降低隧道应变。Wang等(2012)通过数值分析,发现覆盖有缓冲层的隧道,能量耗散率有所降低。

    山岭隧道地震响应特征相较于地下隧道有着显著的差异(Wang et al,2001Lin et al,2020),地震波在山体内会发生反射及相干效应,山体、隔震层、隧道作为“散射体”和“二次震源”可改变地震动分布与数值,在研究时需要考虑山岭场地的地震动放大效应及山体-隔震层-隧道之间的动力相互作用。目前对于山岭内隧道隔震措施的研究多基于模型试验(梅松华等,2022杨长卫等,2023),而针对基于复杂山岭场地地形的双线隧道隔减震研究还有待深入。

    综上,为探究隔减震措施对山体双线隧道地震波散射的影响规律,本文采用无奇异间接边界元法(Ba,Yin,2016Fang et al,2016),拟对“山体场地-隔震层-双线隧道”系统地震的进行动整体响应分析。基于对一系列参数的分析,定性分析了双线山岭隧道在地震作用下的空间分布规律,并定量揭示了入射波特性、隔减震措施属性(如材料的弹性模量和厚度)等参数对隧道地震动响应的影响,以期山岭隧道的减隔震设计和施工提供部分参考。

    图1a所示,高斯形山体内穿越有双线衬砌隧道,在隧道与山体之间设置有隔震层。假定山体场地、隧道和隔震层均为线性、各向同性介质。几何参数定义如下:L1表示水平地表边界,L2表示高斯山体表面,L3L5分别表示左和右隧道衬砌内边界,L4L6分别表示左和右隔震层域外边界,L7L8分别表示左和右隧道衬砌外边界;D1代表弹性半空间域,D2D3代表隧道域,D4D5代表隔震层域;设高斯山体底半径为a,高为h,设左、右两隧道的内半径和外半径分别为r1r2,隔震层外半径为r3,衬砌厚度为t,两隧道净距为d0,隧道中心与地表距离为hd。假设平面P波从基岩半空间中以角度θ入射(与竖向夹角)。

    图  1  山体内双圆形衬砌隧道计算模型(a)和错动滑移边界构造(b)
    L3L5分别表示左右隧道衬砌内边界;L7L8分别表示左右隧道衬砌外边界;D2D3代表隧道域;r1r2为隧道内外半径;r3为隔震层外半径;t为衬砌厚度,下同
    Figure  1.  Calculation model of double circular lining tunnel in the mountain (a) and staggered slip boundary structure (b)
    L3 and L5 are the inner boundaries of the left and right tunnels,respectively;L7 and L8 are the outer boundaries of the left and right tunnels,respectively;D2 and D3 represent tunnel domains;r1 and r2 are the inner and outer radii of the tunnel,respectively;r3 is the outer radius of the isolation layer;t is the thickness of the lining,the same below

    考虑到地震动力作用下,隧道与围岩之间不是完全固结,存在一定的错动滑移,在隔震层与围岩之间设置一系列虚拟的线性弹簧和阻尼器,以模拟它们之间的不完美边界(图1b)。同时,假设接触力与相对位移成正比,即认为该模型力学状态处于弹性阶段,未考虑弹塑性、大变形等非线性效应(Yi et al,2014Fang et al,2016)。

    根据弹性波动理论,总波场可分为自由场和散射场的叠加。半空间域D1同时受到自由场和散射场作用(Huang et al,2019),隧道域D2D3与隔震层域D4D5内只受到散射场作用,各域划分如图2所示。根据单层位势理论和间接边界元法(indirect boundary element,缩写为IBEM )原理,散射场引起的位移u和应力$\delta $的积分表达式如下:

    图  2  隧道—山体系统域划分及边界离散
    (a) 弹性半空间域D1;(b) 隔震层域D4D5;(c) 隧道域D2D3
    Figure  2.  Domain-division of the tunnel-hill system and boundary discrete elements
    (a) Elastic half-space domain D1;(b) Seismic isolation layer domain D4D5);(c) Tunnel domain D2D3
    $$ u_i^{{\mathrm{s}}} ( x ) = \int_S {{\phi _j} ( \xi ) } {G_{ij}} ( x \text{,} \xi ) {\mathrm{d}}{S_\xi }\text{,} $$ (1)
    $$ \sigma_i^{\mathrm{s}} ( x ) =-0.5\delta_{ij}\phi_j ( x ) +\int_S^{ }\phi_j ( \xi ) T_{ij} ( x\text{,} \xi )\mathrm{ d}S _{\xi}, $$ (2)

    式中: $ {\phi _j}{\mathrm{d}}{S _\xi } ( i \text{,} j = x \text{,} y ) $为离散边界单元的应力,${\mathrm{d}{S _\xi }} $为散射场引起的位移;${G_{ij}} ( x \text{,} \xi ) $和${T_{ij}} ( x \text{,} \xi ) $分别代表位移和应力格林函数,表示波源点ξj方向上的单位力,在x点引起i方向的位移或应力,格林函数自动满足无限远辐射条件,同时满足波动方程。

    1) 弹性半空间域D1在水平地表L1和山体表面L2上满足牵引力为零,则边界条件为:

    $$ \sigma _{{D_{1 \text{,} ij}}}^{{\mathrm{f}}} + \sigma _{{D_{1 \text{,} ij}}}^{{\mathrm{s}}} = 0 { \text{.}}$$ (3)

    2) 半空间域D1与隔震层域D4D5在边界L4L6上存在错动滑移,满足应力连续、位移不连续,即

    $$ u_{{D_{1 \text{,} {\mathrm{n}}}}}^{\rm f} + u_{{D_{1\text{,} {\mathrm{n}}}}} ^{\rm s} - u_{{D_{4 ( 5 ) \text{,} \mathrm{n}}}}^{\rm s}{=}\frac{{\sigma _{{D_{4 ( 5 ) \text{,} {\mathrm{n}}}}}^{\rm s}}}{{{k_{\mathrm{n}}}}} + {\delta _{\mathrm{n}}}\frac{{ {\text{∂}} \left(u_{{{\mathrm{D}}_{1 \text{,} {\mathrm{n}}}}}^{\rm f} + u_{{{\mathrm{D}_{1 \text{,} {\mathrm{n}}}}}}^{\rm s} - u_{{{\mathrm{D_{4 ( 5 ) \text{,} {\mathrm{n}}}}}}}^{\rm s}\right)}}{{ {\text{∂}} {{t}}}} \text{,} $$ (4)
    $$ u_{{D_{1 \text{,} {\mathrm{t}}}}}^{{\mathrm{f}}} + u_{{D_{1 \text{,} {\mathrm{t}}}}}^{\rm s} - u_{{D_{4 ( 5 ) \text{,} {\mathrm{t}}}}}^{\rm s}{=}\frac{{\sigma _{D_{4 ( 5 ) \text{,} {\mathrm{t}}}}^{\rm s}}}{{{k_{\text{t}}}}} + {\delta _{\mathrm{t}}}\frac{{ {\text{∂}} \left(u_{{{\mathrm{D}}_{1\text{,} {\mathrm{t}}}} } ^{\rm f} + u_{{{\mathrm{D}}_{1 \text{,} {\mathrm{t}}}}}^{\rm s} - u_{{{\mathrm{D}}_{4 ( 5 ) \text{,} {\mathrm{t}}}}}^{\rm s}\right)}}{{ {\text{∂}} {{t}}}}\text{,} $$ (5)
    $$ \sigma _{{{D_{1 \text{,} ij}}}}^{{\mathrm{f}}} + \sigma _{{{D_{1 \text{,} ij}}}}^{{\mathrm{s}}} = \sigma _{{{D_{4 ( 5 ) \text{,} ij}}}}^{{\mathrm{s}}} { \text{,}} $$ (6)

    式中:kn,kt分别为法向和切向刚度系数;δn,δt分别为法向和切向黏性系数;定义无量纲刚度系数为k*knr11= ktr11,其中μ1为弹性半空间介质的剪切模量;定义无量纲黏性系数为δ*=δn/r1δt/r1

    3) 隧道域与隔震层域(D2D4D3D5)在边界L7L8上满足应力和位移连续,则边界条件为:

    $$ u_{D_{2 ( {3} ) \text{,} ij}}^{\rm s} = u_{D_{4 ( {5} ) \text{,} ij}}^{\rm s},$$ (7)
    $$ \sigma _{D_{2 ( {3} ) \text{,} ij}}^{\rm s} = \sigma _{D_{4 ( {5} ) \text{,} ij}}^{\rm s} { \text{.}} $$ (8)

    4) 隧道域D2D3在内边界L3L5上满足应力为零,则边界条件为:

    $$ \sigma _{{ D_{2 \text{,} ij}}}^{\rm f} + \sigma _{{D_{2 \text{,} ij}}}^{\rm s}= 0 ,$$ (9)

    边界条件(3)—(9)可表达为以下积分形式:

    $$ \int_S {\phi _j^{D_1} ( \xi ) T_{ij}^{D_1} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} = - \sigma _{D_1}^{\rm f} ; $$ (10)
    $$ \begin{gathered} \int_S {\phi _j^{D_1} ( \xi ) G_{ij}^{D_1} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} - \int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) G_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }}= - u_{D_1}^{\rm f}{ + } \frac{{\int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) T_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} }}{{{k_{\mathrm{n}}}}}; \\ \end{gathered} $$ (11)
    $$ \begin{gathered} \int_S {\phi _j^{D_1} ( \xi ) G_{ij}^{D_1} ( x \text{,} \xi ) {\rm d}{S _\xi }} - \int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) G_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\rm d}{S _\xi }}= - u_{D_1}^{\rm f}{ + } \frac{{\int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) T_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\rm d}{S _\xi }} }}{{{k_{\mathrm{t}}}}} ; \\ \end{gathered} $$ (12)
    $$ \begin{split} & - 0.5[{{\phi _j^{D_1}} ( \xi ) - {\phi _j^{D_{4 ( {5} ) }}} ( \xi ) } ]{ + }\int_S {\phi _j^{D_1} ( \xi ) T_{ij}^{D_1} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} - \\ & \qquad \int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) T_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} = - t_{D_1}^{\rm f} ; \end{split} $$ (13)
    $$ \int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) G_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} - \int_S {\phi _j^{D_{2 ( {3} ) }} ( \xi ) G_{ij}^{D_{2 ( {3} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} = 0 ;$$ (14)
    $$ \begin{split} & - 0.5\left[{{\phi _j}^{D_{4 ( {5} ) }} ( \xi ) - {\phi _j}^{D_{2 ( {3} ) }} ( \xi ) } \right] + \int_S {\phi _j^{D_{4 ( {5} ) }} ( \xi ) T_{ij}^{D_{4 ( {5} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} - \\ & \qquad \int_S {\phi _j^{D_{2 ( {3} ) }} ( \xi ) T_{ij}^{D_{2 ( {3} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} = 0 ; \end{split} $$ (15)
    $$ \int_S {\phi _j^{D_{2 ( {3} ) }} ( \xi ) T_{ij}^{D_{2 ( {3} ) }} ( x \text{,} \xi ) {\mathrm{d}}{S _\xi }} = 0 { \text{.}}$$ (16)

    将边界离散为若干个单元,式(10)—(16)可转化为如下离散形式,式中n1n2轮换,在L1处的离散取n1表达式,在L2处的离散取n2表达式:

    $$ \sum\limits_{l = 1}^{{N_1} + {N_2} + {N_4} + {N_6}}{\phi _j^{D_1} ( {\xi _l} ) t_{ij}^{D_1} ( {x_{{n_1}}} \text{,} {\xi _l} ) } = - t_{D_1}^{{\mathrm{f}}} ( {x_{{n_1}}} ) \text{,} \qquad {n_1} ( {{n_2}} ) = 1 \text{,} \cdots \text{,} {N_1} ( {{N_2}} ) ; $$ (17)
    $$ \begin{split} & \sum\limits_{{l_1} = 1}^{{N_1} + {N_2} + {N_4} + {N_6}}{\phi _j^{{D_1}} ( {\xi _{{l_1}}} ) g_{ij}^{{D_1}} ( {x_{{n_4}}} \text{,} {\xi _{{l_1}}} ) } -\sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) } {\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) g_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_4}}} \text{,} {\xi _{{l_2}}} ) } =- u_{{D_1}}^{\rm f}{ + } \\ & \qquad \frac{{\displaystyle\sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) } {\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) t_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_4}}} \text{,} {\xi _{{l_2}}} ) } }}{{{k_{\mathrm{n}}}^{}}} \text{,} \qquad {n_4} ( {{n_6}} ) = 1 \text{,} \cdots \text{,} {N_4} ( {{N_6}} ) ; \end{split} $$ (18)
    $$ \begin{split} & \sum\limits_{{l_1} = 1}^{{N_1} + {N_2} + {N_4} + {N_6}}{\phi _j^{D_1} ( {\xi _{{l_1}}} ) g_{ij}^{D_1} ( {x_{{n_4}}} \text{,} {\xi _{{l_1}}} ) } - \sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) } {\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) g_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_4}}} \text{,} {\xi _{{l_2}}} ) } =- u_{D_1}^{\rm f}{ + } \\ & \qquad \frac{{\displaystyle\sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) } {\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) t_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_4}}} \text{,} {\xi _{{l_2}}} ) } }}{{{k_{\mathrm{t}}}^{}}} \text{,} \qquad {n_4} ( {{n_6}} ) = 1 \text{,} \cdots \text{,} {N_4} ( {{N_6}} ) ; \end{split} $$ (19)
    $$ \begin{split} &- 0.5\left({{\phi _j} ^{D_1} ( \xi ) - {\phi _j}^{D_{4 ( {5} ) }} ( \xi ) } \right)+\sum\limits_{{l_1} = 1}^{{N_1} + {N_2} + {N_4} + {N_6}} {\phi _j^{D_1} ( {\xi _{{l_1}}} ) t_{ij}^{D_1} ( {x_{{n_4}}} \text{,} {\xi _{{l_1}}} )- } \\ & \qquad \sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) }{\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) t_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_4}}} \text{,} {\xi _{{l_2}}} ) } = - t_{D_1}^{\rm f} \text{,} \qquad{n_4} ( {{n_6}} ) = 1 \text{,} \cdots \text{,} {N_4} ( {{N_6}} ) ; \end{split} $$ (20)
    $$ \begin{split} & \sum\limits_{{l_2} = 1}^{{N_4} + {N_7} ( {{N_6} + {N_8}} ) } {\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) g_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_7}}} \text{,} {\xi _{{l_2}}} ) } - \sum\limits_{{l_3} = 1}^{{N_3} + {N_7} ( {{N_5} + {N_8}} ) }{\phi _j^{D_{2 ( {3} ) }} ( {\xi _{{l_3}}} ) g_{ij}^{D_{2 ( {3} ) }} ( {x_{{n_7}}} \text{,} {\xi _{{l_3}}} ) } \\ & \qquad = 0 \text{,} \qquad {n_7} ( {{n_8}} ) = 1 \text{,} \cdots \text{,} {N_7} ( {{N_8}} ) ; \end{split} $$ (21)
    $$ \begin{split}\\ & - 0.5\left[{{\phi _j^{D_{4 ( {5} ) }}} ( \xi ) - {\phi _j^{D_{2 ( {3} ) }}} ( \xi ) } \right] + \sum\limits_{{l_2} = 1}^{{N_4} + {N_7}{{ ( {{N_6} + {N_8}} ) }_{}}}{\phi _j^{D_{4 ( {5} ) }} ( {\xi _{{l_2}}} ) t_{ij}^{D_{4 ( {5} ) }} ( {x_{{n_7}}} \text{,} {\xi _{{l_2}}} ) } - \\ & \qquad \sum\limits_{{l_3} = 1}^{{N_3} + {N_7} ( {{N_5} + {N_8}} ) } {\phi _j^{D_{2 ( {3} ) }} ( {\xi _{{l_3}}} ) t_{ij}^{D_{2 ( {3} ) }} ( {x_{{n_7}}} \text{,} {\xi _{{l_3}}} ) } = 0 \text{,} \qquad {n_7} ( {{n_8}} ) = 1 \text{,} \cdots \text{,} {N_7} ( {{N_8}} ) ; \end{split} $$ (22)
    $$ \sum\limits_{l = 1}^{{N_3} + {N_7} ( {{N_5} + {N_8}} ) } {\phi _j^{D_{2 ( {3} ) }} ( {\xi _l} ) t_{ij}^{D_{2 ( {3} ) }} ( {x_{{n_3}}} \text{,} {\xi _l} ) } = 0 \text{,} \qquad {n_3} ( {{n_5}} ) = 1 \text{,} \cdots \text{,} {N_3} ( {{N_5}} ) { \text{.}}$$ (23)

    上式中,牵引力及位移动力格林函数可表示为:

    $$ {\sigma _{ij}} ( {x_n} \text{,} {\xi _l} ) = \frac{1}{2}{\delta _{ij}}{\xi _{nl}} + \displaystyle\int_{\xi{_ l} - \tfrac{{\Delta S}}{2}}^{\xi {_l} + \tfrac{{\Delta S}}{2}} {{T_{ij}} ( {x_n} \text{,} \xi ) } {\mathrm{d}}{S _\xi }, $$ (24)
    $$ {g_{ij}} ( {x_n} \text{,} {\xi _l} ) = \int_{{\xi _l} - \tfrac{{\Delta S}}{2}}^{{\xi _l} + \tfrac{{\Delta S}}{2}} {{G_{ij}} ( {x_n} \text{,} \xi ) {\mathrm{d}}{S _\xi }}, $$ (25)

    式中:当xξ时,采用高斯积分法来求解$ {\sigma _{ij}} ( {x_n} \text{,} {\xi _l} ) $和$ {g _{ij}} ( {x_n} \text{,} {\xi _l} ) $;当xξ时,为避免格林函数的奇异性,需利用格林函数级数展开式解析求解:

    $$ {\sigma _{ij}} ( {x_n} \text{,} {\xi _n} ) = \frac{1}{2}{\delta _{ij}}, $$ (26)
    $$ {g_{ij}} ( {x_n} \text{,} {\xi _n} ) = - \frac{{i\Delta S}}{{4\mu }}\left[1 + i\frac{2}{\text{π} } ( 1 - {\text{γ }} - \lg \left( \frac{{k\Delta S}}{4}\right) \right] ,$$ (27)

    式中:γ表示欧拉常数(一般取0.5772),ΔS表示边界离散单元长度。

    求解边界积分方程可得虚拟波源密度,进而可得域内各点的散射场。由散射场和自由场叠加即得出总波场(衬砌中仅考虑散射场),从而可得到半空间和衬砌中任意点的位移和应力。

    目前尚未有相关文献给出P波入射下山岭隧道的精确解析解,本文将模型退化为半空间隧道模型并与已发表的结果(Luco,de Barros,1994)进行比较。计算参数定义为:阻尼比ζ=0.001,泊松比ν=1/3,无量纲频率η=0.5,无量纲刚度系数k*=100,无量纲黏性系数δ*=0,剪切波速比为1.0,隧道间距d0=300 m (即两隧道间的地震动相互作用可以忽略)。图3为本文IBEM计算结果与Luco和de Barros (1994)的结果对比,从图中可以看出,本文方法结果与文献结果吻合程度良好,从而验证了本方法精确性。

    图  3  当前模型退化结果同Luco和de Barros (1994)对P波垂直入射所得结果对比
    (a) 地表位移;(b) 衬砌环向应力
    Figure  3.  The comparison of the present results with the degenerated results of Luco and Barros (1994) for vertical incident P waves
    (a) Surface displacement;(b) Lining circumferential stress

    山体围岩、隧道、隔震层材料的参数如表1所示,山体的几何形状可以表示为$y\,=\, a [ -\,2{ ( x/a ) }^{3}\,-\,3{ ( x/a ) }^{2} \,+\, 1 ] \text{,} a < x {\text{≤}} 0; y\,= a [ 2{ ( x/a ) }^{3}- 3{ ( x/a ) }^{2} + 1 ] \text{,} 0 < x < a$,式中:高斯山体底部半径为a=100 m;高为h;左、右隧道衬砌内、外半径均为r1=6.5 m和r2=7.2 m;衬砌厚度为t=0.7 m;隔震层材料选取工程中常用的聚乙烯泡沫厚度为d;黏滞阻尼比取ζ30.02;考虑到高斯山体场地效应和围岩−衬砌之间的错动滑移时,为简化计算,取无量纲刚度系数$k_{{\mathrm{n}}}^{*} $=$k_{{\mathrm{t}}}^{*} $=5,无量纲黏性系数$\delta_{{\mathrm{n}}}^{*} $=$\delta_{{\mathrm{t}}}^{*} $=10 (梁清华等,2018)。

    表  1  材料参数
    Table  1.  Material parameters
    介质 弹性模量
    /MPa
    密度
    /(kg·m−3
    剪切波速
    /( m·s−1
    泊松比
    围岩 5000 2000 1000 0.25
    隧道 32000 2500 2260 0.25
    隔震层① 12 1000 70 0.38
    隔震层② 50 1200 130 0.38
    隔震层③ 100 1400 170 0.38
    下载: 导出CSV 
    | 显示表格

    下面针对隔震层不同材料属性和厚度,对隧道的动应力集中因子(DSCF)进行了定量分析,详细讨论了隔震层弹性模量、厚度、P波入射频率等因素对隧道地震动响应的影响,并对隔震效率进行了评估。其中DSCF为环向应力和入射波应力幅值的比值。

    图4为P波入射不同弹性模量下隔震层的隧道动应力云图。参考实际工程案例,将隔震层厚度为20 cm,无量纲频率分别为1,2,5,10,弹性模量分别取12,50,100 MPa。

    图  4  P波垂直入射下隔震层d和弹性模量$\eta $不同时左隧道的动应力集中因子DSCF
    (a) 无隔震层;(b) 隔震层弹性模量100 MPa;(c) 隔震层弹性模量50 MPa;(d) 隔震层弹性模量12 MPa
    Figure  4.  DSCF of the left tunnel with different elastic modulus of isolation layer under vertical incident P wave
    (a) No isolation layer;(b) Elastic modulus of isolation layer 100 MPa;(c) Elastic modulus of isolation layer 50 MPa;(d) Elastic modulus of isolation layer 12 MPa

    图4可以看出,随着隔震层弹性模量逐渐变小,隧道动应力数值也越来越小,隔震层的隔震效果越好。以η=1时为例,三种隔震层弹性模量100,50,12 MPa对应的峰值动应力集中因子分别为16.9,14.5,10.1,与不设隔震层情况的应力峰值27相比,可分别使应力降低至62.6%,53.7%,37.4%。同时,通过观察应力在隧道截面的分布也可看出,当弹性模量较低时,除了可以降低其数值表现,更可以有效减少其应力放大区的面积,从而有效避免地震作用而导致的衬砌裂缝的产生。

    比较不同入射频率下衬砌应力的数值和分布可以发现,隔震层对于高频波的结构响应的抑制作用更加明显。当入射频率η=10时,弹性模量为12 MPa的隔震层,隧道应力峰值仅为不设隔震层时的25.4%。随着弹性模量的增加,隔震层在高频时的表现会有所减弱,综合其应力分布和数值表现,12 MPa为较为推荐的弹性模量。

    另外,应力主要集中于隧道拱肩位置,这也与实际状况下裂缝产生的位置相符合。因此在设置隔震层为隔减震措施时,建议也对隧道拱肩区域进行二次加固,以增强隔震效果。

    图5给出了P波在垂直入射时,设置不同隔震层厚度的隧道DSCF分布,根据3.2.1节研究结果,取隔震层模量为12 MPa (在几组弹性模量中隔震效果最好),无量纲频率分别为η=1,2,5,10,隔震层厚度分别取h=10 cm,20 cm和40 cm。

    图  5  P波垂直入射下不同隔震层厚度d和弹性模量$\eta $时DSCF
    (a) 隔震层厚度10 cm;(b) 隔震层厚度20 cm;(c) 隔震层厚度40 cm
    Figure  5.  DSCF of the left tunnel with different shock absorption layer thickness under P wave incidence
    (a) Isolation layer thickness 10 cm;(b) Isolation layer thickness 20 cm;(c) Isolation layer thickness 40 cm

    图5可以看出,隔震层具有明显的隔震效果,随着隔震层厚度的增大,隧道衬砌应力逐渐减小。如垂直入射(η=1),左隧道无隔震层时(图4a)峰值动应力集中因子为27.0,而隔震层厚度为10 cm,20 cm和40 cm时DSCF分别为13.8,10.1和8.6,与不设置隔震层时相比,分别减小约51%,37%和32%,隧道应力降低显著。在几组不同隔震层厚度情况中,40 cm厚度下隔震效果最好。

    总体上看,随着入射频率的增大,衬砌应力放大系数减小,但应力分布更加复杂。但隔震层的存在可使衬砌沿环向的应力分布趋于均匀,应力集中效应降低,使衬砌的整体结构性能得到更好的发挥。在频率变化方面,通过对比不同隔震层厚度在频率η=1和10时的动应力集中因子峰值,三个厚度下的隔震效率分别为54.3%,54.5%和45.5%,可以看出在20 cm厚度下隔震的效率最高。

    因此,综合考虑隔震效率以及工程成本等情况,隔震层的弹性模量设为12 MPa、厚度为20 cm的综合效果较优。

    图6给出了P波垂直入射时隧道加速度峰值(peak ground acceleration,缩写为PGA),取隔震层弹性模量为12 MPa,选取的隧道点位分别为隧道拱顶、拱底和左右拱腰,隔震层厚度分别取10 cm,15 cm和20 cm。把不设隔震层时对应点位的加速度时程作为对照组,可以看出,总体上不同观测点的PGA与不设隔震层时相比有所降低,说明隔震层的设置在降低山岭场地中双线隧道的动力响应上的科学性。

    图  6  P波垂直入射下不同隔震层厚度的左隧道加速度时程(不设隔震层时的加速度时程为参照,如灰色线条所示)
    (a) 隔震层厚度为0 cm与10 cm对比图;(b) 隔震层厚度为0 cm与15 cm对比图;(c) 隔震层厚度为0 cm与20 cm对比图
    Figure  6.  Acceleration time history of the left tunnel with different seismic isolation layer thickness under P wave incidence (taking the acceleration time histories without isolation layer as reference,which are denoted by gray curves)
    (a) Comparison of isolation layer thickness 0 cm and 10 cm;(b) Comparison of isolation layer thickness 0 cm and 15cm;(c) Comparison of isolation layer thickness 0 cm and 20 cm

    本文将IBEM推广到山体场地中双线衬砌隧道地震反应分析,定量研究了P波入射下,聚乙烯泡沫材料的隔震层对山体隧道地震响应规律,研究结果可为山岭场地中双线隧道的隔减震设计和施工提供部分参考。得到以下主要结论:

    1) IBEM可以精确求解山体场地中衬砌隧道群的地震动力反应,包括山岭地震动放大效应、衬砌的应力集中效应等,设置隔减震措施能有效改变应力分布,并充分发挥围岩的承载能力,从而起到保护隧道衬砌的作用。

    2) 随着隔震层材料的弹性模量减小,隧道动应力数值有明显的减小。柔性隔震层对于高频波结构响应的抑制作用更加明显,高频时弹性模量为12 MPa的隔震层,隧道应力峰值降低到不设隔震层时的25.4%。同时,当弹性模量较小时,可以有效减少其应力放大区的面积,从而有效避免地震作用导致的衬砌裂缝。

    3) 隔震层可使衬砌环向应力分布趋于均匀,且随着隔震层厚度的增大,隧道衬砌动应力减小,减小幅度可约达50%以上。

    4) 综合考虑隔震效率以及工程成本等情况,根据本文研究的结果,隔震层的弹性模量设为12 MPa、厚度为20 cm时的综合效果较优。

    5) 考虑地震动力作用下隧道与围岩之间存在碎土,在边界上引入了错动滑移边界。本文中提出的错动滑移边界模型为简化模式,未考虑弹塑性、大变形等非线性效应。

  • 图  1   山体内双圆形衬砌隧道计算模型(a)和错动滑移边界构造(b)

    L3L5分别表示左右隧道衬砌内边界;L7L8分别表示左右隧道衬砌外边界;D2D3代表隧道域;r1r2为隧道内外半径;r3为隔震层外半径;t为衬砌厚度,下同

    Figure  1.   Calculation model of double circular lining tunnel in the mountain (a) and staggered slip boundary structure (b)

    L3 and L5 are the inner boundaries of the left and right tunnels,respectively;L7 and L8 are the outer boundaries of the left and right tunnels,respectively;D2 and D3 represent tunnel domains;r1 and r2 are the inner and outer radii of the tunnel,respectively;r3 is the outer radius of the isolation layer;t is the thickness of the lining,the same below

    图  2   隧道—山体系统域划分及边界离散

    (a) 弹性半空间域D1;(b) 隔震层域D4D5;(c) 隧道域D2D3

    Figure  2.   Domain-division of the tunnel-hill system and boundary discrete elements

    (a) Elastic half-space domain D1;(b) Seismic isolation layer domain D4D5);(c) Tunnel domain D2D3

    图  3   当前模型退化结果同Luco和de Barros (1994)对P波垂直入射所得结果对比

    (a) 地表位移;(b) 衬砌环向应力

    Figure  3.   The comparison of the present results with the degenerated results of Luco and Barros (1994) for vertical incident P waves

    (a) Surface displacement;(b) Lining circumferential stress

    图  4   P波垂直入射下隔震层d和弹性模量$\eta $不同时左隧道的动应力集中因子DSCF

    (a) 无隔震层;(b) 隔震层弹性模量100 MPa;(c) 隔震层弹性模量50 MPa;(d) 隔震层弹性模量12 MPa

    Figure  4.   DSCF of the left tunnel with different elastic modulus of isolation layer under vertical incident P wave

    (a) No isolation layer;(b) Elastic modulus of isolation layer 100 MPa;(c) Elastic modulus of isolation layer 50 MPa;(d) Elastic modulus of isolation layer 12 MPa

    图  5   P波垂直入射下不同隔震层厚度d和弹性模量$\eta $时DSCF

    (a) 隔震层厚度10 cm;(b) 隔震层厚度20 cm;(c) 隔震层厚度40 cm

    Figure  5.   DSCF of the left tunnel with different shock absorption layer thickness under P wave incidence

    (a) Isolation layer thickness 10 cm;(b) Isolation layer thickness 20 cm;(c) Isolation layer thickness 40 cm

    图  6   P波垂直入射下不同隔震层厚度的左隧道加速度时程(不设隔震层时的加速度时程为参照,如灰色线条所示)

    (a) 隔震层厚度为0 cm与10 cm对比图;(b) 隔震层厚度为0 cm与15 cm对比图;(c) 隔震层厚度为0 cm与20 cm对比图

    Figure  6.   Acceleration time history of the left tunnel with different seismic isolation layer thickness under P wave incidence (taking the acceleration time histories without isolation layer as reference,which are denoted by gray curves)

    (a) Comparison of isolation layer thickness 0 cm and 10 cm;(b) Comparison of isolation layer thickness 0 cm and 15cm;(c) Comparison of isolation layer thickness 0 cm and 20 cm

    表  1   材料参数

    Table  1   Material parameters

    介质 弹性模量
    /MPa
    密度
    /(kg·m−3
    剪切波速
    /( m·s−1
    泊松比
    围岩 5000 2000 1000 0.25
    隧道 32000 2500 2260 0.25
    隔震层① 12 1000 70 0.38
    隔震层② 50 1200 130 0.38
    隔震层③ 100 1400 170 0.38
    下载: 导出CSV
  • 李天斌. 2008. 汶川特大地震中山岭隧道变形破坏特征及影响因素分析[J]. 工程地质学报,16(6):742–750. doi: 10.3969/j.issn.1004-9665.2008.06.003

    Li T B. 2008. Failure characteristics and influence factor analysis of mountain tunnels at epicenter zones of great Wenchuan earthquake[J]. Journal of Engineering Geology,16(6):742–750 (in Chinese).

    梁清华,刘中宪,周晓洁,何颖. 2018. 基于粘性—滑移界面接触模型半空间隧道衬砌对平面P波的散射[J]. 世界地震工程,34(3):111–123.

    Liang Q H,Liu Z X,Zhou X J,He Y. 2018. Scattering of plane P waves by a lined tunnel in a half-space based on viscous-slip interface contact model[J]. World Earthquake Engineering,34(3):111–123 (in Chinese).

    梅松华,盛谦,崔臻,梅贤丞. 2022. 黏弹性阻尼减震层的吸能特性试验研究[J]. 岩土工程学报,44(6):997–1005.

    Mei S H,Sheng Q,Cui Z,Mei X C. 2022. Experimental study on energy absorption property of viscoelasticity damping layer[J]. Chinese Journal of Geotechnical Engineering,44(6):997–1005 (in Chinese).

    杨长卫,张良,张凯文,岳茂,童心豪,温浩. 2023. 山岭隧道跨断裂带段及洞口段地震响应大型振动台模型试验研究[J]. 岩石力学与工程学报,42(4):993–1002.

    Yang C W,Zhang L,Zhang K W,Yue M,Tong X H,Wen H. 2023. Large scale shaking table model test on seismic response of mountain tunnel portal section passing through fault zone[J]. Chinese Journal of Rock Mechanics and Engineering,42(4):993–1002 (in Chinese).

    禹海涛,李晶,王祺. 2022. 软土隧道基于地震响应的输入地震动排序[J]. 地震学报,44(1):123–131. doi: 10.11939/jass.20210166

    Yu H T,Li J,Wang Q. 2022. Ranking the seismic input motion based on seismic response of soft soil tunnels[J]. Acta Seismologica Sinica,44(1):123–131 (in Chinese).

    Ba Z N,Yin X. 2016. Wave scattering of complex local site in a layered half-space by using a multidomain IBEM:Incident plane SH waves[J]. Geophys J Int,205(3):1382–1405. doi: 10.1093/gji/ggw090

    Bobet A. 2009. Elastic solution for deep tunnels:Application to excavation damage zone and rockbolt support[J]. Rock Mech Rock Eng,42(2):147–174. doi: 10.1007/s00603-007-0140-0

    Fang X Q,Jin H X,Liu J X,Huang M J. 2016. Imperfect bonding effect on dynamic response of a non-circular lined tunnel subjected to shear waves[J]. Tunnell Undergr Space Technol,56:226–231. doi: 10.1016/j.tust.2016.03.008

    Huang L,Liu Z X,Wu C Q,Liang J W. 2019. The scattering of plane P,SV waves by twin lining tunnels with imperfect interfaces embedded in an elastic half-space[J]. Tunnell Undergr Space Technol,85:319–330. doi: 10.1016/j.tust.2018.12.024

    Jiang X L,Wang F F,Yang H,Sun G C,Niu J Y. 2018. Dynamic response of shallow-buried small spacing tunnel with asymmetrical pressure:Shaking table testing and numerical simulation[J]. Geotech Geol Eng,36(4):2037–2055. doi: 10.1007/s10706-017-0444-0

    Lin Z,Yan L,Xiang C,Yang H Y. 2020. Study on dynamic response laws and shock absorption measures of mountain tunnel under strong earthquake[J]. Adv Civil Eng:1671838.

    Luco J E,de Barros F C P. 1994. Dynamic displacements and stresses in the vicinity of a cylindrical cavity embedded in a half-space[J]. Earthq Eng Struct Dyn,23(3):321–340. doi: 10.1002/eqe.4290230307

    Ma X,Wang F M,Guo C C,Sun B. 2020. Seismic isolation effect of non-water reacted two-component polymeric material coating on tunnels[J]. Appl Sci,10(7):2606. doi: 10.3390/app10072606

    Meng G W,Gao B,Zhou J M,Cao G D,Zhang Q. 2016. Experimental investigation of the mechanical behavior of the steel fiber reinforced concrete tunnel segment[J]. Constr Build Mater,126:98–107. doi: 10.1016/j.conbuildmat.2016.09.028

    Su L J,Liu H Q,Yao G C,Zhang J L. 2019. Experimental study on the closed-cell aluminum foam shock absorption layer of a high-speed railway tunnel[J]. Soil Dyn Earthq Eng,119:331–345. doi: 10.1016/j.soildyn.2019.01.012

    Wang W L,Wang T T,Su J J,Lin C H,Seng C R,Huang T H. 2001. Assessment of damage in mountain tunnels due to the Taiwan Chi-Chi earthquake[J]. Tunnell Undergr Space Technol,16(3):133–150. doi: 10.1016/S0886-7798(01)00047-5

    Wang Z,Zhu W,Zhang Z. 2012. Seismic isolation effect of a tunnel covered with a buffer layer[J]. Dis Adv,5(4):1372–1376.

    Xu X Y,Wu Z J,Sun H,Weng L,Chu Z F,Liu Q S. 2021. An extended numerical manifold method for simulation of grouting reinforcement in deep rock tunnels[J]. Tunnell Undergr Space Technol,115:104020. doi: 10.1016/j.tust.2021.104020

    Yi C P,Zhang P,Johansson D,Nyberg U. 2014. Dynamic response of a circular lined tunnel with an imperfect interface subjected to cylindrical P-waves[J]. Comput Geotech,55:165–171. doi: 10.1016/j.compgeo.2013.08.009

    Zhou X H,Cheng X S,Qi L,Wang P,Chai S F,Liu Y J. 2021. Shaking table model test of loess tunnel structure under rainfall[J]. KSCE J Civ Eng,25(6):2225–2238. doi: 10.1007/s12205-021-1064-z

图(6)  /  表(1)
计量
  • 文章访问数:  140
  • HTML全文浏览量:  37
  • PDF下载量:  38
  • 被引次数: 0
出版历程
  • 收稿日期:  2023-08-13
  • 修回日期:  2023-11-12
  • 网络出版日期:  2024-06-25
  • 刊出日期:  2024-05-14

目录

/

返回文章
返回