Abstract Summary
Title: Identification of a Coupled Ground-Building Seismic Response Analysis Model Based on Measured Data In recent years, unexpected large earthquakes have caused much damage, and it has become very important to know the vibration characteristics of buildings. In particular, medium- and low-rise buildings are often seen throughout the world. It is especially necessary to properly evaluate this effect for structural design, because these buildings in general are strongly affected by the dynamic interaction effect between the ground and the building. On the other hand, it is not always the case that the building after completion of construction or the ground at the point of construction will have the dynamic characteristics assumed in the design. In order to verify their consistency and provide basic information during design, inverse analysis based on observation records is helpful. For this purpose, it is desirable to identify not only the natural period and damping constants, but also the components of the dynamic analysis model, such as the stiffness of each floor for the lumped mass analysis model. However, most of the previous studies have focused on building models with fixed foundations, and there have been few cases of identification for coupled ground-building systems. In addition, most of these studies have been conducted on shaking table experiments, but there are also few studies on actual buildings, where observation points are limited. In this paper, we apply the Modal Iterative Error Correction method (MIEC method) to the earthquake records of the foreshocks of the 2011 off the Pacific coast of Tohoku earthquake to identify the coupled superstructure-foundation-soil system in a real building with a small number of observation points. The analysis was performed on an eight-story SRC building. Observations were made on the first, second, fifth, and eighth floors, and this data was used for the identification. In this earthquake, the response of each part of the building is within a linear range. The advantages of using the MIEC method include the following; 1) A large number of parameters can be identified at the same time. 2) It is not prone to locally stable solutions. As an identification method, multiple parameters are identified in two steps by dividing them into each building model. First, the stiffness of the superstructure is identified as a parameter in the fixed-foundation model. Next, an SR model was created using the stiffness of the superstructure as the initial value, and the identification was performed with the ground spring as a parameter. The identification results for a real building with a limited number of observation points including soil springs in a linear range showed good correspondence with the observed values, and the identification was performed appropriately.
