Sufficiency assessments of ground motion intensity measures employing kullback-leibler theory (applied for typical south pars offshore platforms)

The potential ingrained uncertainty in ground motion records may significantly influence the structural seismic risk assessment in performance-based earthquake engineering (PBEE). One of the basic components of the socio-economic method of PBEE design is probabilistic seismic demand model (PSDM).  The level of uncertainty in PSDM, depends greatly on the selected seismic intensity measure (IM), while these models are traditionally conditioned on a single IM. Among various terms utilized in optimal IM selection, this study particularly aims to bring the “sufficiency” assessment procedures into focus. However, the IM efficiency evaluations have also been considered. The sufficiency of IM is gauged by the extent to which the residual demand measure values are statistically independent of ground motion magnitude (Mw) and distance (R), regressing of IM. The objective of this study is to introduce a recently emerged quantitative procedure by employing relative sufficiency measure (RSM) on the basis of Kullback‐Leibler divergence concepts to indicate the superiority of one IM relative to another in the representation of ground motion uncertainty. Besides, the traditional methods of sufficiency evaluation are also discussed. To this end, a three-dimensional finite element model of typical South Pars fixed pile-founded offshore platforms has been built. Several IM candidates are classified and compared in terms of the expected difference in the information they provide for predicting a wide range of structural response parameters. It can be deduced that the most informative of the fourteen considered IMs are among velocity-related ones. The results also demonstrate the absolute necessity of the RSM in optimal IM ranking.