Abstract: The statistical analysis on the dead load of ten domestic suspension bridges with spans exceeding one kilometer was carried out, and the statistical parameters of various self-weight dead load random variables were obtained. The self-weight dead load effect of suspension bridges was divided into three parts such as main cable self-weight effect, stiffening girder self-weight effect and bridge deck system self-weight effect. The calculation formulas of partial safety factors for each dead load were derived on the basis of the optimization method, and the iterative algorithm was established. By calculating and analyzing various combinations of different live-to-dead load ratios and different dead load proportions, the valuing ranges of partial safety factors for each dead load and the equivalent partial safety factor for all dead loads were obtained. The research results indicate that the live-to-dead load ratio has a relatively small impact on the partial safety factors of the dead load, while the proportion of each dead load has a relatively large impact. The value of equivalent partial safety factors for all dead loads ranges from 1.04 to 1.10, and the upper limit value of the partial safety factor for dead load given in the current suspension bridge specifications is 1.10. When the proportion of main cable dead load is large and the proportion of bridge deck system dead load is small, the equivalent partial safety factor takes the lower limit value. Conversely, when the proportion of main cable dead load is small and the proportion of bridge deck system dead load is large, the equivalent partial safety factor takes the upper limit value. Taking a double-deck steel truss girder suspension bridge with a main span of 1860m as an example, its equivalent partial safety factor for dead load is 1.065, which can save 4.43% of the main cable wire area, compared to the partial safety factor for dead load of 1.10 given in the specification.

Key words: bridge engineering; suspension bridge; partial safety factor; degree of reliability; probability distribution

摘要： 对国内多座跨径超过千米的悬索桥恒载进行统计分析，得到了各项自重恒载随机变量的统计参数；将悬索桥自重恒载效应拆分为主缆自重效应、加劲梁自重效应和桥面系自重效应等3部分，基于优化方法推导出各项恒载分项系数计算公式，并建立了迭代算法；通过对不同活恒比及不同恒载占比的多种组合计算分析，获得了各项恒载分项系数及全部恒载等效分项系数的取值范围。研究结果表明：活恒比对恒载分项系数的影响较小，而各项恒载占比对恒载分项系数的影响相对较大；大跨度悬索桥的全部恒载等效分项系数取值为1.04~1.10，现行悬索桥规范中给定的恒载分项系数1.10为上限值；当主缆恒载占比大、桥面系恒载占比小时，其等效分项系数取下限值；当主缆恒载占比小、桥面系恒载占比大时，其等效分项系数取上限值。以某主跨1 860 m的双层桥面钢桁架梁悬索桥为例，其等效恒载分项系数为1.065，相比于规范给定的恒载分项系数1.10，可节约主缆钢丝面积4.43%。

关键词: 桥梁工程；悬索桥；分项系数；可靠度；概率分布