中文摘要：

振动台试验是研究和评价结构抗震性能的主要方法之一。在振动台试验中，按相似理论设计的缩尺模型，通过一系列特定顺序地震动的输入，激励起模型结构的动力反应，来研究和评价结构的抗震性能。对于进入非线性的模型结构来说，振动台试验是个不可逆过程。因此如何选择地震动，以及如何确定地震动的输入顺序，对振动台试验成功与否，显得尤为重要。本文提出以主要周期点处地震动反应谱的包络值与设计反应谱相差不超过20％的方法选择地震动；按主要周期点处多向地震动反应谱的加权求和值大小确定地震动输入顺序；最后将该方法应用到八层可再生混凝土框架模型振动台试验中。试验结果表明，按本文方法所选的地震动可以合理地评价结构的抗震行为，地震动输入顺序可以逐步激励起结构的动力反应。

英文摘要：

Shaking table test is one of the major methods to study and evaluate the seismic poll, tures. In a shaking table test, the scaled model designed by using the theory of similarity is excited by the selected earthquake ground motions in a specified sequence to excite the dynamic responses, which are used to study and e- valuate the structural seismic performance. Due to the nonlinear history of the model, a shaking table test is an irre- versible process. Then how to select the ground motion input and determine the input sequence is very important for the success of the shaking table test. This paper proposes that the ground motion inputs could be selected by setting the difference between the envelop value of the response spectra of earthquake ground motions at major structural periods and the value of the design response spectrum to be less than 20% of the value of the design response spec- trum. The input sequence could be determined according to the weighted sum of the response spectra of earthquake grond motions in multi-directions at major structural periods. The shaking table test of an 8-story renewable rein- forced concrete frame is applied to verify the method. The experimental results show that the seismic performance of the structure could be reasonably evaluated by the selected ground motions and the dynamic responses could be ex- cited according to the input sequence determined by the proposed method in this paper.