Journal of Seismology and Earthquake Engineering
Volume:16 Issue: 1, Spring 2014

The purpose of this study is to develop a model for a class of unbounded domains with application in infinite beams on elastic supports. An unphysical layer is included in the model in order to absorb the crossing waves into the unbounded domain. To this end, the Perfectly Matched Layer (PML) is used along with a displacement-based Finite Element scheme that provides an appropriate vehicle for such problems. Most PML applications appearing in the literature have dealt with lower order governing differential equations. The case of a beam on elastic foundation, on the other hand, involves a fourth-order equation. The governing equation is reduced into a series of four first-order equations by introducing auxiliary variables. This leads to internal moments and shear forces that represent non-linear behaviour in the artificial medium. The accuracy of PML results is validated by comparison with regular finite element solutions of beams with substantially long spans. The solution method is used to investigate dynamic response of railroad tracks under earthquake excitations. The effect of various parameters on seismic response and the resonance phenomenon has been examined. Numerical results demonstrate the accuracy and efficacy of the method, which is due to use of small bounded domains in the solution process.

The seismic linear demand of structures is usually reduced by employing a forcereduction factor inmost of force-based seismic regulations. The current force reduction factors in ASCE/SEI 41-06 standard results in conservative design in the case of foundations when compared to the conventional design regulations e.g. IBC 2000. The aim of the current paper is to evaluate the influence of the Soil-Foundation-Structure-Interaction (SFSI) effects on the force-reduction factor based on ASCE/SEI 41-06 standard. Therefore, a comparison has been made between the results of the nonlinear response history analysis of a set of 3, 6, 10 and 15-storeyed concrete 2-D frames with the result of the equivalent linear static approach. The results show that the equivalent linear static approach load combinations, in the case of foundations, can lead to conservative designs. Finally, a set of new force-reduction factor has been calculated in order to cope with this problem.

Optimal placement of dampers with proper characteristics is the major concern for seismic rehabilitation of structures using these passive devices. In this paper, endurance time method is used to find the optimum placement of dampers in steel frames at different seismic hazard levels. Genetic algorithm is combined with ET method to overcome this optimization problem. Different types of nonlinear modeling are used to validate ETanalysis results. ETacceleration functions used in this study are generated based on ASCE7-05 design spectrum. Results of ETanalysis are compared with the results of time history analysis using seven near-field ground motions spectrally matched to the same spectrum. The Maximum difference between the results of two methods is about 15% in this study, and results of ET analysis are compatible with the results of time history analysis at different hazard levels. Consistency of the results shows that ET method can be used as an effective procedure to optimize damper placement in structures reducing the required computational effort.

This paper is devoted to present a method of three-dimensional stability analysis of concave slopes in plan view based on lower-bound theorem of the limit analysis approach in static and seismic cases. Slope stability problems are often analyzed two-dimensionally by conventional limit equilibrium method (LEM). Accuracy of LEM is often questioned due to the underlying assumptions that it makes, and at the same time, analyzing a 3D problem two-dimensionally may lead to significant differences in safety factors depending on the slope geometries. In this paper, the numerical linear finite element and the rigorous lower bound limit analysis method is used to produce some seismic stability dimensionless charts for three-dimensional (3D) homogeneous concave slopes. The charts can be used by practicing engineers as a convenient tool to estimate the stability for excavated or man-made slopes. The results obtained using this 3D method showthat the stability of concave slopes increases as the relative curvature R/H and the relative width of slope decrease.

The generation of riskmaps in urban area is necessary for risk and disaster management studies and strategic planning. This research is completed in line with the tasks of the GEM-EMME (Global Earthquake Model - Earthquake Model of the Middle East) project and aims at the development of a systematic procedure in estimating the seismic risk to residential buildings and the human loss. The procedure involves four main stages. At first, the geospatial information for buildings and population are compiled and processed as to generate a country-wide geodatabase. In the next phase, ground shaking maps are produced for scenario or actual earthquakes for case studies. The third phase corresponds to the vulnerability modeling of the building stock. In the final phase, the estimate of the building damage and the associated casualty loss are calculated for two case studies. For the first case study, the residential building loss and casualty numbers are calculated for three important earthquake scenarios produced by Mosha Fault (MF), North Tehran Fault (NTF) and Rey Fault (RF) for Tehran. The current findings show that NTF can potentially account for the high number of 349428 heavily damaged or destroyed housing units and 100680 severely injured (or dead) people outnumbering MF and RF cases. For Rey fault, the estimated figures correspond to 257329 heavily damaged (or destroyed) housing units and 54468 severely injured (or dead). For the second case study, 7760 heavily damaged (or destroyed) housing units and 1045 severely injured (or dead) people are estimated for Ahar-Varzeghan earthquake (August 2012) that is in close agreement with the actual reported numbers.

Correlations between seismic acceleration parameters and damage indices can help to predict the value of the damage for an earthquake event. This paper has two parts. One part is the detection of the interdependency between important seismic acceleration parameters, and two damage models including the modified Park-Ang cumulative damage model and the maximummodified flexural damage ratio model (MFDR) by using of the Spearman correlation coefficient. For the first part, we have utilized 17 records of the earthquake from all over the world. Results showed sustained maximum acceleration and effective design acceleration have the best correlation with the Park-Ang model. Besides, the weakest interdependencies are related to displacement RMS and cumulative absolute velocity for the Park-Ang model. However, among selected seismic acceleration parameters, peak ground velocity and Housner Intensity have the best interdependencies with the MFDR model. On the other hand, cumulative absolute velocity and Vmax/Amax have shown the weakest interdependencies with the MFDR model. Other part includes the time variation of the MFDR model in duration of the Victoria earthquake. It can give a good sight about process of behavior members during earthquake. Moreover, time variation of the MFDR model can determine process degradation of each member.