Experimental and computational verification of reasonable design formulae for base‐isolated structures

Abstract

SUMMARYIt is clear that base isolation is a sensible strategic design in attenuating the responses of a structural system induced by ground motions. The design of seismically isolated structures is mainly governed by the Uniform Building Code (UBC) published by the International Conference of Building O cials. The UBC code emphasizes a simple, statically equivalent design method that displacements of an isolated structure are concentrated at the isolation level. Therefore, the superstructure nearly moves as a rigid body and the design forces of elements above isolators are based on the behaviour of isolators at the design displacement. However, in the UBC code, the distribution of inertial (or lateral) forces over the height of the superstructure above isolation has been found to be too conservative for most isolated structures. In view of this, two simple and reasonable design formulae for the lateral force distribution on isolated structures have been proposed in this paper. Results obtained from a full-scale isolated structure tested on the shaking table and numerical analyses of two additional examples verify the suitability of design formulae. It is illustrated that the proposed formulae can predict well the lateral force distribution on isolated structures during earthquakes.

Lateral force distributions for the linear static analysis of base-isolated buildings

“…The comparison is carried out for the 5-storey building, endowed with two different IS's, whose mechanical characteristics are: (a) F y /W = 0.05, D y ∼ = 5 mm, T 1 = 0.6 s, α = 5%, µ ∼ = 30 (corresponding to ξ is ∼ = 24% and NL ∼ = 0.38) and (b) F y /W = 0.05, D y = 10 mm, T 1 = 0.9s, α = 5%, µ ∼ = 20 (corresponding to ξ is ∼ = 29% and NL ∼ = 0.46). These IS parameters are fully compatible with those selected by other authors in their studies (Tsai et al 2003;Ryan and York 2007;Andriono and Carr 1991b).…”

“…Such discrepancies can be basically ascribed to three main aspects: (i) different assumptions in the IS modelling, (ii) different criteria in the selection of the input ground motions and (iii) different approach in the examination of the NTHA results. In the studies by Tsai et al (2003) and Lee et al (2001), for instance, reference is made to an equivalent viscous-elastic model, which could be suitable for HDRB-IS (as explicitly recognized by the authors), but not for IS's characterised by a strong nonlinear behaviour or large values of the effective damping (e.g. LRB, FPB, etc.).…”

“…Recently, two different groups of researchers (Tsai et al 2003;Lee et al 2001) proposed two similar modified equations for the vertical distribution of the equivalent static forces, considering the influence of the first mode of vibration of the BI-structure. The proposed equations can be expressed as follows:…”

“…H e represents the effective height of the equivalent SDOF system of the BI-structure, considering the flexibility of the superstructure. The main difference between the equations proposed by Lee et al (2001) and Tsai et al (2003) is that, with reference to framed structures, the former assumes H e = 0.6H whilst the latter assumes H e = H, where H is the total height of the superstructure. In essence, the force distribution given by Eq.…”

On the Distribution in Height of Base Shear Forces in Linear Static Analysis of Base-Isolated Structures

“…Therefore, enhanced formulations are also available in the literature to describe the distribution in height of lateral forces with the aim to take into account relevant parameters, such as the frequency ratio between BI-and fixed-base structure [41,42], the fundamental period of the superstructure [43], the modal shape of the latter [42], the non-linearity factor [14] of seismic isolators [44,45] or combination of some of the above parameters [46][47][48][49]. Under the mathematical standpoint, the proposed formulations are based on either introducing an exponent to h i in Equation (2) [44,46] or combining Equations (1) and (2) [41,42,48].…”