DESIGN METHOD OF SEISMICALLY ISOLATED STRUCTURES BASED ON GENERALIZED RESPONSE SPECTRUM ANALYSIS

Abstract

 Seismic isolation is widely used in seismic areas of the world for their functions of securing both of human life and the property. While simple design methods for simple structures have been proposed by many researchers, these methods are generally limited to specific seismically isolated structural systems. Particularly in Japan, the number of challenging structural systems (e.g. a base isolation with tall buildings, mid-story isolation or roof isolation systems having substructures with dampers) where these methods cannot be applied are recently increasing. In practice, many projects require an iterative approach with time-consuming NLRHA to determine the isolation design. Therefore, this paper presents a design method for highly indeterminate seismically isolated structures utilizing generalized response spectrum analysis, which has been extended to simulate both of an elasto-plastic damper with isolator and a nonlinear oil damper. In section 2, the extension of generalized response spectrum analysis (GRSA) is described in detail. The design method of seismically isolated structures based on GRSA is proposed. In section 3, the fundamental accuracy of GRSA is verified using a series of numerical simulation of single-story isolated structure models. Moreover, the range in application of the parameter formulations of complex stiffness for complex multi shear spring element simulating elastoplastic dampers with isolator is discussed. In section 4, the design example of a tall building with base isolation based on GRSA is demonstrated. In section 5, the design example of a 9-story mid-story isolation building having substructure with dampers based on GRSA is demonstrated. In section 6, the design example of an isolated lattice dome building having substructure with dampers based on GRSA is demonstrated.

 In summary, the following results were obtained:

 1) GRSA is extended for highly indeterminate seismically isolated structures and their accuracy was verified in a series of comparison studies with non-linear response history analyses.

 2) While modified geometrical stiffness method (Mod-GSM) is suitable for seismically isolated structure with shorter initial natural period (T₀ = 0.1 and 0.3, ) average damping method (ADM) is suitable for seismically isolated structure with longer initial natural period (T₀ = 0.7 and 1.0.) The range in application of ADM to keep the evaluation error within 30% is that ductility ratio is lower than 1000 and the nonlinear coefficient NL is lower than 0.6. The range in application of Mod-GSM to keep the evaluation error within 30% is that the nonlinear coefficient NL is lower than 0.7.

 3) A design method of seismically isolated structures based on the contour map of GRSA is proposed and the efficiency is demonstrated using a series of the design examples.

 4) When additional dampers are distributed in supporting substructures below the isolated layer of mid-story isolated structures or seismically isolated roof structures, the horizontal stiffness of the substructure incorporating elasto-plastic damper braces should be more reduced compared with that of the substructure incorporating oil damper braces for reducing the acceleration response of the substructure. However, the response acceleration of the substructure may not be reduced for the above seismically isolated systems having substructure with elasto-plastic damper braces.