Search

The main objectives of this article are to develop a technique to find source models that allow one to replicate observed strong ground motion records and to extrapolate strong ground motion synthesis to locations where strong motion was not recorded. A technique including the well known empirical Green's function (EGF) approach along with a genetic algorithm is used, which allows the optimization of differences between the synthesized and observed ground shakings. The technique used is performed by comparing the elastic response spectra of observed seismograms at two stations with those of simulated data using the EGF method incorporating recorded aftershocks taken at each station. Moreover, a genetic algorithm approach is used to reduce differences between the simulated and recorded data in the form of elastic response spectra by changing the input parameters in the admissible ranges. To validate the proposed approach the three components of strong motion recorded at other stations were synthesized incorporating the input parameters obtained at previous stations. A comparatively good match of the simulated and recorded response spectra confirms the ability of the proposed technique to generate synthetic seismograms with suitable elastic response spectra. Les principaux objectifs de cet article sont de developper une technique pour trouver des modeles de sources qui permettent de dupliquer les enregistrements de forts mouvements du sol observes et d'extrapoler la synthese des forts mouvements du sol a des endroits ou le fort mouvement n'a pas ete observe. Une technique qui comprend l'approche par la fonction empirique bien connue de Green et un algorithme genetique est utilisee; elle permet l'optimisation des differences entre les tremblements du sol observes et synthetises. La technique utilisee compare le spectre de la reponse elastique des sismogrammes observes a deux stations avec ceux des donnees simulees en utilisant la methode de la fonction empirique de Green incorporant des repliques sismiques a chaque station. De plus, une methode d'algorithme genetique est utilisee pour reduire les differences entre les donnees simulees et enregistrees sous la forme de spectre de reponse elastique en changeant les parametres d'entree dans les plages admissibles. Afin de valider l'approche proposee, les trois composantes des forts mouvements enregistres aux autres stations ont ete synthetisees en incorporant les parametres d'entree obtenus anterieurement a d'autres stations. La concordance assez bonne entre les spectres de reponse simules et enregistres confirme la capacite de la technique proposee pour generer des sismographes synthetiques avec un spectre de reponse elastique adequat. [Traduit par la Redaction], Introduction The main objectives of this article are to develop a technique to find source models that allow one to replicate observed strong ground motion records and to extrapolate strong [...]

Progressive approaches to the incorporation of directivity effects into the probabilistic seismic hazard analysis require a reliable model to be able to predict the pulse observing probability. This study proposes a new model that predicts the probability of observing pulse-like ground motion at a site, using the source rupture characteristics as the predictor variables. As a result, a new 3D source-to-site geometry is used to define a directivity parameter. The resulting predictive model can be incorporated into the near-fault seismic hazard analysis. This proposed prediction model is validated by comparing the estimated results with those observed at near-field stations during previous events. These comparative results have shown that the proposed model agrees well with the available near-fault ground-motion data from different earthquake scenarios. [ABSTRACT FROM AUTHOR]

One of the most important challenges in seismic evaluation of structures comes from wide variability of possible motions that a specific site may experience. This variability arises from the intrinsic randomness of seismic motions (aleatory uncertainty) and lack of knowledge about the physical process behind an earthquake (epistemic uncertainty). The epistemic uncertainty in estimating ground motion can be properly characterized by providing an understanding of the physical aspects of earthquake source models. In this paper, a combination of the theoretical-based Green's function method and hybrid slip source model is used to reduce the source uncertainty. The procedure is implemented on the 1994 Northridge earthquake as a case study. For this purpose, the method is validated against recordings stations through demonstrating a good agreement between the elastic response spectra corresponding to the simulated and recorded data, confirming the reliability of the procedure. A simple additive weighting algorithm is used to select the best fit amongst simulated waveforms and optimal source model. [ABSTRACT FROM AUTHOR]

SUMMARY Tall buildings are long-period structures that are sensitive to the long-period content of ground motions. Selection of appropriate ground motions is an important step in seismic collapse assessment of tall buildings using nonlinear dynamic analyses. Epsilon ( ε Sa) and eta ( η) are two spectral shape indicators, which have been recently proposed for ground motion selection in the technical literature. In this study, a new parameter gamma ( γ) is proposed, which has considerable correlation with the collapse capacity of long-period structures having a fundamental period greater than 1 s. This parameter is a linear combination of ε Sa and the displacement spectrum intensity epsilon ( εDSI). The parameter γ is obtained and optimized by applying the particle swarm optimization algorithm. Since γ has significant correlation with the collapse capacity of long-period structures, it can be used as an efficient proxy for ground motion selection in seismic collapse assessment of tall buildings. The results of this study show that ground motion selection considering the new proxy γ causes reduction in the dispersion of structural response and also decrease in the mean annual frequency of collapse, when compared with ground motion selection based on ε Sa and η. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Pulse-like near-fault ground motions, which are characterized by the presence of a velocity pulse, can impose large demands in structures, and hence can potentially increase seismic collapse risk. One of the most important parameters affecting the structural demands under near-fault ground motions is the pulse period. The commonly used scalar intensity measure (IM), spectral acceleration at the fundamental period of the structure, Sa(T1) is demonstrated to be deficient, and also insufficient with respect to the pulse period, for predicting the structural collapse capacity under pulse-like ground motions. Furthermore, it is shown that the recently proposed IM, named INP, is not able to fully account for the effect of the pulse period. This study proposes an optimal integral-based IM, named ISa, for reliable seismic collapse assessment of structures subjected to pulse-like near-fault ground motions. ISa is the integral of the pseudo acceleration response spectrum over an optimal period range that is a function of the fundamental period and ductility of the structure. To propose the new IM, 15 generic frame structures with different fundamental periods and ductility classes were employed. The results indicate that ISa is significantly efficient, which causes reduction in the dispersion of structural collapse capacity, and also sufficient with respect to the pulse period and other ground motion characteristics for seismic collapse capacity assessment. [ABSTRACT FROM AUTHOR]

Reliable use of b-value in log [N (M)] = a -- bM is critical in seismicity comparisons, seismic hazard analysis, prediction and comparative mechanism studies. Since earthquakes and the b-value are considered as stochastic processes and random variable, respectively, applying the probability distribution of b-value is necessary in its temporal and spatial variations assessment. In this paper, we propose a novel method that employs the b-value uncertainty in probabilistic seismic hazard analysis using normal-exponential joint distribution function. To this end, we calculate b-value statistics based on bootstrap sampling of the seismic catalog. Our analytical and experimental evaluations show that the proposed joint distribution results in a more precise closed-form relation for the probabilistic seismic hazard analysis accurately reducing the hazard in comparison to conventional methods. The benefit of the proposed approach here is improving the ability of assessing the effectiveness of various seismic risk mitigation strategies and so, allocates the available resources more efficiently. [ABSTRACT FROM AUTHOR]

Recurrence models apply historical seismicity information to seismic hazard analysis. These models that play an important role in the obtained hazard curve are determined by their parameters. Recurrence parameters estimation has some features that lie in missing-data problems category. Thus, the observed data cannot be used directly to estimate model parameters. Furthermore discussion about results reliability and probable conservatism is impossible. The present study aims at offering an approach for Gutenberg-Richter parameters (a and b-values) estimation and determine their variation. Applying the proposed method to analyses of the heterogeneous data sets of seismic catalog, one would calculate valid estimates for recurrence parameters. This method has the capability to reflect all known sources of variability. The results of the case study clearly demonstrate applicability and efficiency of the proposed method, which can easily be implemented not only in advanced but also in practical seismic hazard analyses. [ABSTRACT FROM AUTHOR]

This study is intended to investigate the contribution of fling step to seismic demands of the selected structures. Due to the scarcity of real data recorded at the two sides of the causative fault including fling step effects, Fandoqa scenario (Iran, 1998) is simulated using a theoretical Green's function. The three components of the seismograms at Sirch station are simulated using the developed Multi-Objective Particle Swarm Optimization algorithm (MO-PSO) algorithm. The model is calibrated by incorporating the obtained optimal model parameters. The seismograms at four stations close to and located both sides and the causative fault are generated and used as the input data. The fling step contributions are removed from the synthetic data thus providing the seismograms without fling effects. Two structures, a 3D-30 story steel structure dual system (MRF and X-brace) and a 2D-20 story special R/C frame are selected. These two structures are dynamically analyzed under synthetic waveforms with and without fling step contributions. The nonlinear seismic demands over the height of the selected structures with and without fling step contribution are calculated and assessed. It is concluded that in general, fling step may increase or decrease the seismic demands and is not a predictable problem. [ABSTRACT FROM AUTHOR]

Mass irregular structures have stories with effective mass more than 150% of effective mass of adjacent storey. Seismic response of mass irregular structures are assessed by consideration of SMRF structures with 4, 6 and 8 stories, which mass irregularity is applied separately by 1.4 and 2 times mass changes in top and two intermediate stories. Seismic response is achieved by applying Incremental Dynamic Analysis. Mean annual frequency and probability of exceedance in 50 years of performance levels are evaluated by applying PBEE framework. Finally understood which mass irregular structures response is related to location and number of stories with mass changes. Also, 150% limitation of mass change for definition of mass irregularity need more study. [ABSTRACT FROM AUTHOR]

This article is intended to predict a seismogram at sites other than those from which the model is calibrated using the two well-known stochastic point-source and finite-fault approaches. The two models are implemented on the 2005 Zarand earthquake as a case study. For this purpose, the models are validated against recordings at four stations through demonstrating a good agreement between the elastic response spectra corresponding to the synthesized and recorded data, confirming the reliability of the two models. Additionally, a quality factor and maximum cut-off frequency are proposed for the Zarand region (Iran). It is concluded that the models can be used to predict seismogram for engineering purposes in the Zarand region. [ABSTRACT FROM PUBLISHER]