【摘要】 大跨度斜拉桥黏滞阻尼器的参数设计通常基于特定地震波的单一构件性能优化，未考虑地震动随机性的影响，以及构件间的相对损伤情况。针对上述问题，提出基于易损性和响应面的参数优化方法：以一座大跨斜拉桥为例，采用OpenSees建立非线性动力模型，基于易损性分析理论评估不同黏滞阻尼器参数下的斜拉桥抗震性能，利用响应面拟合阻尼器参数与易损性的非线性函数关系，以系统易损性最小、损伤路径合理为目标进行参数优化。结果表明：速度指数一定时，随着阻尼系数的增大，各支座的抗震性能单调递增，桥塔和斜拉索的抗震性能先增后减，过渡墩和辅助墩的抗震性能变化很小，基本不受阻尼器参数的影响；斜拉桥黏滞阻尼器的最优参数是桥塔与支座抗震性能综合最优的结果；未优化构件损伤路径所得参数的系统抗震性能更优，但损伤路径并不合理，会高估斜拉桥的实际抗震表现。更多还原

【Abstract】 Parametric design of viscous damper for long-span cable-stayed bridge is usually based on performance optimization of a single component under a specific seismic wave without considering effects of randomness of ground motions and relative damage among components. Here, aiming at the above problems, a parametric optimization method based on vulnerability and response surface was proposed. Taking a large-span cable-stayed bridge as an example, its nonlinear dynamic model was established using the software OpenSees, and its aseismic performance under different viscous damper parameters was evaluated based on the vulnerability analysis theory. Nonlinear function relations among damper parameters and vulnerability were fitted using response surface, and parametric optimization was performed with the target of minimizing system vulnerability and reasonable damage path. The results showed that when the velocity index is constant, the aseismic performance of each support monotonically increases with increase in damping coefficient, while aseismic performances of bridge towers and stay cables firstly increase and then decrease, aseismic performances of transition piers and auxiliary piers change very little and are basically not affected by damper parameters; the optimal parameters of viscous damper for cable-stayed bridge are the comprehensively optimal results of aseismic performances of bridge towers and supports; the system’s aseismic performance may be better if its parameters are obtained from the unoptimized components’ damage path, but this damage path is not reasonable, the actual aseismic behavior of cable-stayed bridge can be overestimated.更多还原