Strategy for the selection of input ground motion for inelastic structural response analysis based on naïve Bayesian classifier - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Bulletin of Earthquake Engineering Année : 2015

Strategy for the selection of input ground motion for inelastic structural response analysis based on naïve Bayesian classifier

Résumé

An application of the naïve Bayesian classifier for selecting strong motion datain terms of the deformation probably induced on a given structural system is presented. Themain differences between the proposed method and the “standard” procedure based on theinference of a polynomial relationship between a single intensity measure and the engineeringdemand parameter are: the discrete description of the engineering demand parameter; theuse of an array of intensity measures; the combination of the information issued from thetraining phase via a Bayesian formulation. Six non-linear structural systems with initialfundamental frequency of 1, 2 and 5 Hz and with different strength reduction factors aremodelled. Their behaviour is described using the Takeda hysteretic model and the engineeringdemand parameter is expressed as the relative drift. A database of 6,373 strong motion recordsis built from worldwide catalogues and is described by a set of “classical” intensity measures;it constitutes the “training dataset” used to feed the Bayesian classifier. The structural systemresponse is reduced to a description of three possible classes: elastic, if the induced driftis lower than the yield displacement; plastic, if the drift ranges between the yield and theultimate drift values; fragile if the drift reaches the ultimate drift. The goal is to evaluate theconditional probability of observing a given status of the system as a function of the intensitymeasure array. To validate the presented methodology and evaluate its prediction capability,a blind test on a second dataset, completely disjointed from the training one, composed of7,000 waveforms recorded in Japan, is performed. The Japanese data are classed using theprobability distribution functions derived on the first data set. It is shown that, by combiningseveral intensity measures through the likelihood product, a stable result is obtained whereby most of the data (>75 %) are well classed. The degree of correlation between the intensitymeasure and the engineering demand parameter controls the reliability of the probabilitycurves associated to each intensity measure.
Fichier non déposé

Dates et versions

hal-03505503 , version 1 (30-12-2021)

Identifiants

Citer

Maria Lancieri, M. Renault, D. Baumont, Catherine Berge-Thierry, Philippe Gueguen, et al.. Strategy for the selection of input ground motion for inelastic structural response analysis based on naïve Bayesian classifier. Bulletin of Earthquake Engineering, 2015, 13 (9), pp.2517-2546. ⟨10.1007/s10518-015-9728-z⟩. ⟨hal-03505503⟩
27 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More