Validating synthetic accelerograms to assess structural seismic risk

Flavia De Luca¹, Iunio Iervolino¹, Gabriele Ameri^2, Francesca Pacor²

¹Università degli Studi di Napoli Federico II

Via Claudio, 21, 80125 Napoli

flavia.deluca@unina.it, iunio.iervolino@unina.it

²Istituto Nazionale di Geofisica e Vulcanologia

Via Bassini, 15, 20133 Milano

ameri@mi.ingv.it, pacor@mi.ingv.it

KEYWORDS: strong-motion simulation, synthetic seismograms, earthquake scenario, IDA, fragility curve.

TOPIC - B  - Vulnerabilità e Rischio Sismico

ABSTRACT

Synthetic accelerograms are an attractive option to assess seismic structural performance in those cases when suitable real records are not available to represent a scenario of interest. On the other hand, to date, it is not exhaustively proven that synthetics allow to an unbiased estimation of structural performance. This is also because it is not straightforward to establish a reference or point of comparison. In the presented study nonlinear structural response, for single and multiple degree of freedom systems (SDOFs and MDOFs), to synthetic accelerograms is compared to a benchmark comprised of real records. Comparison is carried out in terms of fragility or seismic risk.

Synthetic accelerograms are calculated with physics-based simulation methods that accounts for the kinematic characteristics of the rupture propagation over an extended fault plane and for the wavefield attenuation from the source to a specific site.

Synthetic accelerograms are generated for two different earthquake scenarios (magnitude, distance, soil type, etc.). Input data for the definition of the scenario are strictly correlated with the simulation technique selected. A point of comparison (POC) is then evaluated for categories of engineering demand parameters (EDP), the one representative of the peak responses and one representative of the cyclic structural responses. The POC approach (PEER report,2009) allows the definition of a ground motion prediction equation (GMPE) in terms of EDP in which it is possible to get the "true" response of the specific structure given the earthquake scenario. The POC is then employed to investigate whether a generation procedure of synthetic accelerograms is suitable as input for time-history analyses. The EDP-hazard curve obtained by means of the POC for the given scenario is then compared with the one calculated using synthetic records to assess whether they lead to comparable risk assessment.

References

PEER ground motion selection and modification working group. Haselton CB, editor. Evaluation of Ground Motion selection methods: prediction median interstory drift response of buildings. PEER report 2009/01. 2009. Available at http://peer.berkeley.edu/publications/peer_reports/reports_2009.