中文摘要：

为使试验结果能准确地反映原型的抗震性能，对模型的设计方法进行研究，分别按以承载力相似新方法和配筋率相同的常用设计方法设计模型构件，进行伪静力试验，把试验结果与原型构件试验结果进行对比分析；结果表明：等配筋率的设计方法易高估原型结构的力学性能，而承载力相似设计方法的试验结果误差较小。针对振动台试验中采用逐级加载方式给试验结果带来的误差，提出分级相似关系，即在每级次地震动输入后测得模型频率，根据自振频率比和弹性模量比之间的关系，来改变下阶段试验的弹性模量比，从而调整下一阶段相似关系来减小误差；且应用分级相似关系进行振动台试验，将试验结果与不采用分级相似关系的试验结果及直接对模型进行一次性输入的试验结果进行对比，验证分级相似关系的可行性，并提出分级相似关系的适用范围。

英文摘要：

In order to make model tests accurately reflect the seismic performance of the prototype, the method of model design is studied. Static tests were conducted on two groups of components： one is designed by the new design method in which the load on the model and the prototype is in accordance with the similitude law, and the other is designed by the usual method in which the reinforcement ratio of the model and the prototype is equal. Comparison of the model test results with the prototype test results shows that the usual design method overestimates the mechanical properties of prototype while the new design method can make the test result closer to the prototype test. In order to decrease the test result error caused by gradually loading during the shaking table test, a grading similitude relationship is proposed, that to measure the natural frequency of the model after loading in every step, and then to change the elastic modulus ratio between the model and the prototype in the next step according to the relationship between the natural frequency ratio and the elastic modulus ratio. Shaking table test is carried out following the grading the result with the test result when the load is exerted on the element in relationship is verified. The scope of application of the grading similitude similitude relationship, and by comparing one step, the feasibility of grading similitude relationship is given.