In current structural analyses on segmental linings, the bending stiffness of segmental joints is usually treated as a constant, which results in a deviation in the calculated internal forces and deformation. In view of this, a finite element model of segmental linings is established using one ring and two half rings. An iterative algorithm is designed with a 3D curved surface of bending stiffness for the segmental joint, obtained from a series of full-scale tests under different axial force and bending moment cases. The rational convergence criteria for the iterative algorithm are recommended by evaluating the convergence efficiency of the iterative algorithm. Moreover, the influence of load magnitude and initial value of bending stiffness, as well as the changing laws of structural internal forces and deformation with the assembly angles, are presented and analyzed. The results show that with the same load, the convergence of the iterative algorithm in staggered-jointed assembly structures (STGS) is better than that of straight-jointed assembly structures (STRS), owing to its greater overall stiffness. The initial input value of bending stiffness has no effect on the convergence of the proposed iterative algorithm, and generally the convergence performance in STGS is better than that in STRS. In STRS, the bending moment is most sensitive to the changes of assembly angles, while in STGS the assembly angles have great influence on both the bending moment and structural displacement. Without the stiffness iteration, the maximum differences for maximum axial force, bending moment, and displacement under different assembly angle in STRS are 11.7%, 31.9% and 22.3%, respectively, and that differences in STGS are 10.4%, 59.3% and 35.1%, respectively. With the stiffness iterative method, the increase amplitude of the maximum axial force, the maximum bending moment, and the maximum displacement in STRS and STGS are 50.0%, 317.4%, 77.1% and 41.9%, 459.3%, 156.5%, respectively, indicating that changing the assembly angles is a rational way to control the internal structural forces and deformations.

在当前的管片衬砌结构分析中，管片节点的抗弯刚度通常被视为一个常数，这导致计算的内力和变形出现偏差。有鉴于此，采用一个环和两个半环建立了管片衬砌有限元模型。通过在不同轴力和弯矩情况下的一系列全尺寸试验，为节段接头设计了具有弯曲刚度的 3D 曲面的迭代算法。通过评估迭代算法的收敛效率，推荐迭代算法的合理收敛准则。此外，荷载大小和抗弯刚度初始值的影响，以及结构内力和变形随装配角度的变化规律，进行了介绍和分析。结果表明，在相同载荷下，交错连接装配结构（STGS）迭代算法的收敛性优于直连接装配结构（STRS），因为其整体刚度更大。弯曲刚度的初始输入值对提出的迭代算法的收敛性没有影响，一般来说，STGS 中的收敛性能优于 STRS 中的收敛性能。在STRS中，弯矩对装配角度的变化最为敏感，而在STGS中，装配角度对弯矩和结构位移的影响都很大。在不进行刚度迭代的情况下，STRS 中不同装配角度下的最大轴力、弯矩和位移的最大差异分别为 11.7%、31.9% 和 22.3%，STGS 的差异分别为 10.4%、59.3% 和 35.1%。采用刚度迭代法，STRS和STGS中最大轴力、最大弯矩和最大位移的增加幅度分别为50.0%、317.4%、77.1%和41.9%、459.3%、156.5%，表明改变装配角度是控制内部结构力和变形的合理方法。