（1.中铁大桥勘测设计院集团有限公司，湖北 武汉 430050； 2.武汉青山长江大桥建设有限公司，湖北 武汉 430345）
摘 要：武汉青山长江公路大桥主桥为主跨938 m的钢箱及钢箱结合梁斜拉桥，中跨主梁为钢箱梁，采用分节段悬臂施工至跨中合龙。为保证成桥后的内力及线形满足设计要求，通过3D-Bridge有限元软件建立主桥计算模型，基于无应力状态法，确定合理目标状态，并对钢箱梁制造、架设、合龙，以及斜拉索施工控制技术进行研究。施工过程中，按照合龙及成桥目标状态的主梁线形、斜拉索索力进行施工控制；通过钢箱梁纵、横向制造线形控制及厂内预拼，减少制造线形误差；钢箱梁节段间采用先栓接后焊接的连接方式，控制施工线形；基于合龙口影响因素敏感性分析结果，确定主动合龙的方式；考虑斜拉索非线性效应，对主梁线形及索力进行双控。控制结果表明：主梁线形偏差小于10 cm，索力偏差小于10%，满足设计要求。
中图分类号： U448.27；U445.4 文献标志码： A
Key Construction Control Techniques for Central-Span Steel Box Girders of Qingshan Changjiang River Highway Bridge in Wuhan
LIAO Gui-xing1, YAN Ru-hui2, HU Hui-yue1, ZHANG Yan-fei1, XU Gong-yi1
(1. China Railway Major Bridge Reconnaissance & Design Institute Co., Ltd., Wuhan 430050, China; 2. Wuhan Qingshan Changjiang River Bridge Construction Co., Ltd., Wuhan 430345, China)
Abstract: The main bridge of Qingshan Changjiang River Highway Bridge in Wuhan is a cable-stayed bridge with a main span of 938 m. The superstructure consists of steel box girders and steel box-concrete composite girders, among those, the steel box girders in the central span are erected by segmental cantilever assembly method and closed in midspan. To ensure that the internal forces and geometry of the bridge in the completed state could meet the design requirements, the finite element software 3D-Bridge was used to establish the computing model of the main bridge. The rational target geometry of the bridge was determined based on the stress-free state method, and the specific construction control techniques were studied, including the manufacturing, erection and closure of the steel box girders as well as the installation of stay cables. The longitudinal and transverse manufacturing geometries of the steel box girders were controlled, and the components were preassembled in the factory, to reduce the manufacturing geometry errors. The steel box girder segments were bolted together first, and then welded, to control the construction geometry. An active way of closure was selected based on the sensitivity analysis of factors influencing the closure gap. Given the existence of the non-linearity effect of stay cables, both the main girder geometry and the cable forces were controlled. The results of the construction control demonstrate that the geometry error of the main girder is less than 10 cm, and the cable force error is less than 10%, meeting the design requirements.
Key words: cable-stayed bridge; steel box girder; stress-free state method; target condition; geometry; cable force; active closure; construction control