PROBABILISTIC EVALUATION OF SEISMIC PERFORMANCE FOR A STEEL MOMENT FRAME USING DAMAGE INDICES

Abstract eng:
Reliable predictions of seismic structural damages are of essential importance in earthquake engineering. In order to make proper and effective decisions in post-earthquake situations, it is necessary to have realistic estimations of the extent of seismic damages occurred in individual structures. For this purpose, structural damage indices can be very useful. Defined in a variety of forms, they can serve as strong analytical tools needed to calculate different aspects of structural damages caused by earthquakes. This study aims to provide a comparison among some of the well-known analytical damage indices, within the scope of seismic performance evaluations, and using a probabilistic framework. A sample 5-storey special moment-resisting frame is modeled using OpenSees. Damping, seismic mass, yield strength and ultimate strength of steel are described as probabilistic variables, and random realizations of the structure are generated using Latin Hypercube Sampling (LHS) method. Sample frames are then analyzed through Extended IDA. Increasing levels of damage, due to increasing levels of strong ground motion intensity, are described with the selected set of seismic damage indices. Considering the results, each damage index displays a different pattern for IDA curves based on the response parameters used in it. Seismic performance levels, as defined in some seismic design criteria, are correlated to the order of damage indices’values. Also, statistical dispersion in a set of damage values corresponding to a particular performance level is different for each index, indicating different levels of sensitivity to epistemic uncertainties among damage indices.

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