Conventional validation of analytical and numerical models in Earthquake Engineering involves the comparison of numerically simulated response time histories to experimentally obtained benchmark responses to the same earthquake excitations. As the seismic design problem is inherently stochastic, an alternative, statistical, and easier-to-pass validation procedure has been suggested. As an example, numerical and analytical models may fail to predict the planar rocking response of a rigid block to a specific ground motion, but they can be proven quite successful in predicting the statistical distribution of the maxima of that response to an ensemble of ground motions. This article describes the publicly available data obtained from a series of 226 shake table tests of a 3D rocking podium structure, designed at ETH Zurich and carried out at EQUALS Lab, University of Bristol. This well-documented dataset is the largest one involving a shake table and can be used to statistically validate analytical and numerical models of rocking structures.

地震工程中分析模型和数值模型的常规验证涉及将数值模拟的响应时间历史记录与通过实验获得的对相同地震激励的基准响应进行比较。由于地震设计问题本质上是随机的，因此提出了一种替代的，统计的和易于通过的验证程序。例如，数值和分析模型可能无法预测刚性块对特定地面运动的平面摇摆响应，但是可以证明它们在预测该响应对地面运动整体的统计分布方面非常成功。 。本文介绍了从一系列226次3D摇摆式讲台结构的振动台测试中获得的公开数据，由苏黎世联邦理工学院设计，并在布里斯托大学的EQUALS实验室进行。这个记录良好的数据集是最大的包含摇床的数据集，可用于统计验证摇摆结构的分析和数值模型。