Evaluation of Ground-Motion Selection & Modification Methods Using Inelastic-Response Hazard Curves

Abstract eng:
Performance assessments of the built environment using numerical simulation uses suites of ground motions which are selected, scaled and modified such that their energy content matches, or exceeds, that of a target spectrum, which is computed from a probabilistic seismic hazard analysis. There are many methods of selecting, scaling and/or modifying ground motions in engineering practice today. However, there is no objective methodology to determine whether the suite of ground motion corresponds to the same level of hazard and/or risk as the target spectrum. This paper presents a through methodology where different ground-motion suites are evaluated using hazard curves (annual rate of exceedance) for inelastic response parameters instead of the elastic models used in the PSHA. Specifically, the paper evaluates three different methods of ground-motion modification: simple amplitude scaling, tight spectral matching (at the record level), and mean spectral matching (at the suite-average level). While tightly-matched records have no dispersion in the elastic spectral response, the records obtained from the other two methods do. Also, the mean of the spectra of the amplitudescaled records are expected to be conservative in comparison to those obtained using the other two methods. These three different methods are, therefore, compared in the inelastic realm on the basis of the mean response of the suite as well as the dispersion of the results. The quantitative measure of comparison is based on probabilistic metrics rather than response measures.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket