为解决近断层地震动作用下双连拱隧道抗震计算中难以同时考虑隧道顶部双拱的几何效应、近断层速度脉冲效应及显著竖向地震效应的问题，建立一种面向近断层区域双连拱隧道横向抗震分析的理论计算方法。首先，针对传统反应位移法中隧道边界位移直接等同于自由场位移、且难以反映双连拱隧道中墙及上覆土体共同作用的力学机制，构建包含双连拱隧道结构及其周围一定范围围岩的广义结构抗震力学模型，明确自由场相对位移、边界剪应力、地基弹簧约束以及结构地震惯性力的施加方式，提出对应的计算流程。其次，基于近断层脉冲型地震动动力时程数值模拟，分析双连拱隧道主洞和中墙峰值加速度与地震动强度参数之间的相关性，建立考虑水平、竖向峰值加速度及埋深修正的结构地震惯性力计算式。同时，通过不同边界范围下理论计算与数值结果差异对比，提出广义结构区域水平向及竖向合理范围的确定方法。然后，采用数值模拟与振动台试验结合的方法对理论方法的有效性进行验证。最后，通过参数分析，揭示结构几何、埋深及地震动参数对地震响应的影响。研究结果表明： 适度增大中墙宽高比可缓解中墙应力集中，但宽高比超过0.20会使衬砌应力大幅增长； 增加埋深能使结构逐步转入以受压为主的有利受力状态； 近断层竖向地震动的增强（PGA_v/PGA_h由0.7增至2.0）可将关键部位应力放大16.7%~130.5%； 速度脉冲效应（PGV/PGA比值）对衬砌外侧和中墙影响显，在0.15~0.27 s区间应力激增明显。

Seismic calculation of double-arch tunnels subjected to near-fault ground motions faces the challenge of simultaneously accounting for the double-arch geometric effect at the tunnel crown, the near-fault velocity pulse effect, and the significant vertical seismic effect. To address this issue, a theoretical calculation method specifically designed for the transverse seismic analysis of double-arch tunnels in near-fault regions is established. First, the limitations of the conventional response displacement method, which directly equates the tunnel boundary displacement to free-field displacement, make it difficult to capture the interactions among the middle wall, overlying soil, and tunnel structure. To overcome this, a generalized structural seismic mechanical model is developed that incorporates the double-arch tunnel structure and a range of surrounding rock. This model defines free-field relative displacement, boundary shear stress, foundation spring constraints, and structural seismic inertial forces, leading to a comprehensive calculation procedure. Next, based on dynamic time-history numerical simulations using near-fault pulse-like ground motions, the relationships among the peak accelerations of the main tunnel, middle wall, and ground-motion intensity parameters are analyzed. From this analysis, calculation formulas for structural seismic inertial forces are established by considering both horizontal and vertical peak ground accelerations along with burial-depth corrections. Based on a comparison of the differences between theoretical calculations and numerical results across various boundary ranges, a method for determining the reasonable transverse and vertical extents of the generalized structural region is proposed. The validity of the proposed theoretical method is confirmed through a combination of numerical simulations and vibration platform tests. Finally, parametric analyses are conducted to explore the effects of structural geometry, burial depth, and ground-motion parameters on structural seismic responses. The results show that a moderate increase in the width-to-height ratio of the middle wall can reduce stress concentration, whereas a ratio exceeding 0.20 may significantly increase lining stress. An increase in burial depth tends to favorably shift the structure toward a compression-dominated stress state. In addition, enhancing near-fault vertical ground motion, with the PGA_v/PGA_h ratio increasing from 0.7 to 2.0, can amplify stresses at critical locations by 16.7% to 130.5%. The velocity pulse effect, represented by the PGV/PGA ratio, significantly influences the outer side of the lining and the middle wall, with stress levels sharply increasing in the interval of 0.15 to 0.27 s.