Keywords: seismic-isolation, bilinear behaviour, equivalent linear model, effective stiffness and damping.

Seismic isolation is used in relatively stiff buildings to reduce the induced seismic loads, by shifting their fundamental periods outside the dangerous for resonance range. Seismic isolation systems exhibit non-linear inelastic behaviour, which often can be sufficiently well represented using bilinear inelastic models. However, simplified linear elastic analysis is frequently used, at least at the preliminary design and analysis phases of seismically isolated buildings. Therefore, it is very important to investigate the appropriateness of using such linearized models, which are proposed in the literature or by relevant design codes and provisions.

The models proposed in the literature, equivalent linearized models that are used to simulate the essentially bilinear hysteretic behaviour of seismic isolation systems, are assessed through parametric studies. The proposed methods and formulas are evaluated by comparing the computed peak responses with those that are computed using the corresponding bilinear inelastic models. In particular, this research work investigates whether linearized models are suitable for the analysis of seismically isolated buildings, considering the maximum relative displacements at the isolation level, the peak inter-storey deflections and the peak absolute floor accelerations.

A number of numerical simulations and parametric studies have been performed, using a specially-developed software, to obtain an insight on the suitability of the proposed in the literature equivalent linear elastic models to simulate multi-storey seismically isolated buildings and identify how that is affected by certain structural parameters and earthquake characteristics. Equivalent linear elastic analyses are conducted using six proposed linearized models, under strong earthquakes ground motions.

The results indicate that the estimation of the effective stiffness shift and the effective viscous damping ratio influences considerably the computed response. The parametric studies show that the peak responses computed by the methods proposed in the literature using linearized models differ, significantly in some cases, from the peak responses computed by the more accurate bilinear inelastic analysis. Since linear elastic analysis, in general, overestimates the maximum relative displacements at the isolation level, it may be used, under certain limitations, in the preliminary design of a seismically isolated building. However, for the analysis of a seismically isolated building and for research purposes relevant to seismic isolation, the more accurate bilinear inelastic analysis should be used, which can be easily performed, nowadays, because of the significantly increased available computing resources.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)