Novel integrated approaches for predicting the compressibility of clay using cascade forward neural networks optimized by swarm- and evolution-based algorithms

Acta Geotechnica - Soft soils are considered as disadvantages in construction, especially in clay layers. It requires many advanced techniques to treat the soft soils before construction, aiming to...     

A hybrid method for the vulnerability assessment of R/C and URM buildings

The methodology followed by the Aristotle University (AUTh) team for the vulnerability assessment of reinforced concrete (R/C) and unreinforced masonry (URM) structures is presented. The paper focuses on the derivation of vulnerability (fragility) curves in terms of peak ground acceleration (PGA), as well as spectral displacement (s _d), and also includes the estimation of capacity curves, for several R/C and URM building types. The vulnerability assessment methodology is based on the hybrid approach developed at AUTh, which combines statistical data with appropriately processed (utilising repair cost models) results from nonlinear dynamic or static analyses, that permit extrapolation of statistical data to PGA’s and/or spectral displacements for which no data are available. The statistical data used herein are from earthquake-damaged greek buildings. An extensive numerical study is carried out, wherein a large number of building types (representing most of the common typologies in S. Europe) are modelled and analysed. Vulnerability curves for several damage states are then derived using the aforementioned hybrid approach. These curves are subsequently used in combination with the mean spectrum of the Microzonation study of Thessaloniki as the basis for the derivation of new vulnerability curves involving spectral quantities. Pushover curves are derived for all building types, then reduced to standard capacity curves, and can easily be used together with the S _d fragility curves as an alternative for developing seismic risk scenarios.     

S-wave spectral analysis of the 1995 Kozani-Grevena (NW Greece) aftershock sequence

S-wave spectral analysis is applied to 174 strong motion accelerationrecords to obtain the source parameters of 27 aftershocks(3.1 M_L 4.3) of the May 13, 1995, M_w 6.6,Kozani-Grevena (NW Greece) earthquake. The data are derived from atemporary network, of three-component digital accelerographs, deployedwithin the strongly affected area some days after the mainshock occurrence.Site effects were evident in the strong motion records at 3 out of the 4stations used, and a correction was applied to account for theoverestimation of seismic moment due to amplification of thelow-frequency part of the spectrum. The data from this analysis arecomplimented with previously obtained source parameters for earthquakesin Greece, in order to study the applicability of the empirical scalingrelations used so far, towards smaller magnitudes. In general, a goodcorrelation was observed in most cases, validating the use of empiricalrelations that are applicable to the Aegean area. Empirical relations aredetermined between seismic moment and seismic slip, as well as, betweenseismic moment and stress drop, applicable to small magnitude earthquakes(M_L < 4.3). Stress drop values were found to be relatively small,ranging from 2 to 41 bars, indicative of inter-plate environments. Thevalues of f_c and of f_max were found in good agreement withrelations based on observations from larger worldwide earthquakes.     

THE DETERMINATION OF FLUORINE BY PROTON INDUCED GAMMA RAY EMISSION SPECTROMETRY (PIGE) IN EIGHTEEN CHINESE STANDARD REFERENCE SAMPLES OF STREAM SEDIMENTS, SOILS AND ROCKS: GSD 9-12, GSS 1-8 AND GSR 1-6

Eighteen new Chinese standard reference samples (including stream sediments, soils and rocks) have been analysed by an automated proton induced gamma ray emission (PIGE) method for fluorine. Results of determinations are reported and are generally in good agreement with the "usable values" previously published.     

Modelling of R/C members accounting for shear failure localisation: Hysteretic shear model

Reinforced concrete (R/C) frame buildings designed according to older seismic codes represent a large part of the existing building stock worldwide. Their structural elements are often vulnerable to shear or flexure‐shear failure, which can eventually lead to loss of axial load resistance of vertical elements and initiate vertical progressive collapse of a building. In this study, a hysteretic model capturing the local shear response of shear‐deficient R/C elements is described in detail, with emphasis on post‐peak behaviour; it differs from existing models in that it considers the localisation of shear strains after the onset of shear failure in a critical length defined by the diagonal failure planes. Additionally, an effort is made to improve the state of the art in post‐peak shear response modelling, by compiling the largest database of experimental results for shear and flexure‐shear critical R/C columns cycled well beyond the onset of shear failure and/or up to the onset of axial failure, and developing empirical relationships for the key parameters defining the local backbone post‐peak shear response of such elements. The implementation of the derived local hysteretic shear model in a computationally efficient beam‐column finite element model with distributed shear flexibility, which accounts for all deformation types, will be presented in a companion paper.     

Modelling of R/C members accounting for shear failure localisation: Finite element model and verification

Reinforced concrete (R/C) frame buildings designed according to older seismic codes represent a large part of the existing building stock worldwide. Their structural elements are often vulnerable to shear or flexure‐shear failure, which can eventually lead to loss of axial load resistance of vertical elements and initiate vertical progressive collapse of a building. In this study, a computationally efficient member‐type finite element model for the hysteretic response of shear critical R/C frame elements up to the onset of axial failure is presented; it accounts for shear‐flexure interaction and considers, for the first time, the localisation of shear strains, after the onset of shear failure, in a critical length defined by the diagonal failure plane. Its predictive capabilities are verified against experimental results of column and frame specimens and are shown to be accurate not only in terms of total response, but also with regard to individual deformation components. The accuracy, versatility, and simplicity of this finite element model make it a valuable tool in seismic analysis of complex R/C buildings with shear deficient structural elements.