10CUEE CONFERENCE PROCEEDINGS 10th International Conference on Urban Earthquake Engineering March 1-2, 2013, Tokyo Institute of Technology, Tokyo, Japan A PRELIMINARY STUDY ON THE IDENTIFICATION OF SEISMIC ISOLATION SYSTEMS FROM EARTHQUAKE RECORDS Giuseppe Oliveto 1) , Anastasia Athanasiou 1) , Mineo Takayama 2) and Keiko Morita 2) 1) Department of Civil and Environmental Engineering, University of Catania, Italy 2)Faculty of Engineering ,Fukuoka University, Japan goliveto@dica.unict.it,athanasiou@dica.unict.it,mineot@fukuoka-u.ac.jp, keikomt@fukuoka-u.ac.jp Abstract: This work is an attempt to the dynamic identification of base isolation systems from earthquake response records. A very simple base isolation system is considered, composed of only four lead rubber bearings at the four corners of a building in the city of Kushiro in northern Japan. The acceleration of the building, recorded during the 2003 Tokachi-oki earthquake, is used for the identification. The identification procedure is based on the minimization of the distance between the acceleration of the building measured during the earthquake and the simulated one obtained by using a model presented at CUEE-12. A state-of-the art optimization algorithm (CMA-ES), based on the adaptation of the covariance matrix of a multivariate Gaussian probability distribution, is used to guide the identification process. The model parameters, obtained by identification from three laboratory tests performed in 1996 and in 2004, are used as a reference for the definition of the search space where the system parameters are expected to be. The optimal solution is given in probabilistic terms, namely the average value, standard deviation and coefficient of variation of the system parameters. The work is completed with a critical analysis of the minimal system response required to obtain a realistic prediction of the system parameters for the model considered. 1. INTRODUCTION The design of seismic isolation systems is usually based on static and/or dynamic laboratory tests performed on individual isolation devices and on suitable models derived from such tests. A model widely used for the simulation of the dynamic response of a wide class of isolation devices, including various types of elastomeric bearings, lead core elastomeric bearings and friction pendulum bearings is the so-called bi-linear spring model, (Naeim and Kelly 1999). Such a model, in parallel with a Coulomb type model for sliding bearings, has been used for the identification of hybrid base isolation systems composed of High Damping Rubber Bearings (HDRB) and Low Friction Sliding Bearings (LFSB) from full scale free vibration tests performed on a base-isolated building, (Oliveto et al. 2004), (Oliveto et al. 2010). The full scale tests are justified by the fact that laboratory tests are usually performed on a limited number of isolation devices, while a full scale test involves the whole isolation system and engages the isolators as they are installed in the construction with possible manufacturing and placing defects. Therefore full-scale tests allow for the assessment of the overall isolation system as it really is. If the test is performed soon after the construction is completed, it could play the role of a certification test that qualifies the isolation system and the building to which it belongs. If it is performed several years after the installation of the isolation system it may be used to assess the present state of the system and to gather some information on the aging process. Free vibration tests on base-isolated buildings have usually been performed by applying a static displacement to the part of the structure above the isolation system and then suddenly removing the force used to apply the initial displacement. The collaborative work by the designer, the constructor, the certifying authority and the owner is required to perform such tests and usually they result in well-conceived and executed experiments. In all known tests, the buildings and the isolation systems were symmetrical, with negligible mass eccentricity or none. This leads to a 1D motion of the superstructure relative to the substructure. For this reason the mechanical models for the identification of the isolation system were 1 Degree of Freedom (1DOF) systems. In spite of the high non-linearity of Hybrid Base Isolation Systems (HBIS) it was possible to provide an analytical solution for the response of such systems under free vibration conditions, (Oliveto et al. 2010), (Athanasiou and Oliveto 2011). The analytical solution made the simulation of the free vibration tests very simple and the identification of the HBIS from such tests very effective. The main ingredients that led to such effectiveness were the analytical solution and the limited number of system parameters to be identified. The least squares method was successfully used for the identification of the isolation system of the Solarino building, (Oliveto et al. 2010). In that case system parameters were identified along with the imposed initial displacement. Subsequent developments allowed for considerable improvements in the simulation of the response of HBIS and in their dynamic identification. A first improvement derived from the extension of the analytical solution from free vibration to forced vibration and earthquake motion, - 1415 -