• Earthquakes and Structures
• /
• v.27 no.1
• /
• pp.41-55
• /
• 2024

There are obvious differences between the characteristics of offshore ground motion and onshore ground motion in current studies, and factors such as water layer and site conditions have great influence on the characteristics of offshore ground motion. In addition, unlike seismic response analysis of offshore superstructures such as sea-crossing bridges, tunnels are affected by offshore soil constraints, so it is necessary to consider the dynamic interaction between structure and offshore soil layer. Therefore, a seismic response analysis model considering the seawater, soil layer and tunnel structure coupling is established. Firstly, the measured offshore and different soil layers onshore ground records are input respectively, and the difference of seismic response under different types of ground motions is analyzed. Then, the models of different site conditions were input into the measured onshore bedrock strong ground motion records to study the influence of seawater layer and silt soft soil layer on the seabed and tunnel structure. The results show that the overall seismic response between the seabed and the tunnel structure is more significant when the offshore ground motion is input. The seawater layer can suppression the vertical seismic response of seabed and tunnel structure, while the slit soft soil layer can amplify the horizontal seismic response. The results will help to promote seismic wave selection of marine structures and provide reference for improving the accuracy of seismic design of immersed tunnels.

• Earthquakes and Structures
• /
• v.17 no.6
• /
• pp.557-566
• /
• 2019

The evaluation of the seismic hazard for a given site is to estimate the seismic ground motion at the surface. This is the result of the combination of the action of the seismic source, which generates seismic waves, the propagation of these waves between the source and the site, and site local conditions. The aim of this work is to evaluate the sensitivity of dynamic response of extended structures to spatial variable ground motions (SVGM). All factors of spatial variability of ground motion are considered, especially local site effect. In this paper, a method is presented to simulate spatially varying earthquake ground motions. The scheme for generating spatially varying ground motions is established for spatial locations on the ground surface with varying site conditions. In this proposed method, two steps are necessary. Firstly, the base rock motions are assumed to have the same intensity and are modelled with a filtered Tajimi-Kanai power spectral density function. An empirical coherency loss model is used to define spatial variable seismic ground motions at the base rock. In the second step, power spectral density function of ground motion on surface is derived by considering site amplification effect based on the one dimensional seismic wave propagation theory. Several dynamics analysis of a curved viaduct to various cases of spatially varying seismic ground motions are performed. For comparison, responses to uniform ground motion, to spatial ground motions without considering local site effect, to spatial ground motions with considering coherency loss, phase delay and local site effects are also calculated. The results showed that the generated seismic signals are strongly conditioned by the local site effect. In the same sense, the dynamic response of the viaduct is very sensitive of the variation of local geological conditions of the site. The effect of neglecting local site effect in dynamic analysis gives rise to a significant underestimation of the seismic demand of the structure.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.2
• /
• pp.119-131
• /
• 2024

Pulse-like ground motion can cause greater damage to structures than nonpulse-like ground motion. Currently, much research is being conducted to determine the presence or absence of velocity pulses and to quantify them from seismic-acceleration records. Existing ground motion is divided into far-field (FF) and near-fault ground motion, based on the distance of the measurement point from the fault. Near-fault ground motion is further classified into near-fault pulse-like (NFP) and near-fault nonpulse-like (NFNP) ground motion by quantifying the presence or absence of velocity pulses. For each ground motion group, 40 FF, 40 NFP, and 40 NFNP ground motions are selected; thus, 120 ground motions are used in the seismic analysis to assess the seismic fragility of sample bridges. Probabilistic seismic demand models (PSDMs) are created by evaluating the seismic responses of two types of sample bridges with lead-rubber and elastomeric rubber bearings using three groups of ground motions. Seismic fragility analysis is performed using the PSDM, and from these results, the effect of the presence or absence of seismic velocity pulses on the seismic fragility is evaluated. From the comparison results of the seismic fragility curve, the seismic fragility of NFP ground motion appears to be approximately three to five times greater than that of NFNP ground motion, according to the presence or absence of a velocity pulse of seismic waves. This means that the damage to the bridge is greater in the case of NFP ground motion than that in the case of NFNP ground motion.

• Earthquakes and Structures
• /
• v.19 no.2
• /
• pp.79-90
• /
• 2020

4- and 8-storey reinforced-concrete frame buildings are analyzed under the suites of the near-fault pulse-like, and the corresponding spectrally equivalent far-fault ground-motion records. Seismic fragility curves for the slight, moderate, extensive, and complete damage states are developed, and the damage probability matrices, and the mean loss ratios corresponding to the Design Basis Earthquake and the Maximum Considered Earthquake hazard levels are compared, for the investigated buildings and sets of ground-motion records. It is observed that the spectrally equivalent far-fault ground-motion records result in comparable estimates of the fragility curve parameters, as that of the near-fault pulse-like ground-motion records. As a result, the derived damage probability matrices and mean loss ratios using two suites of ground-motion records differ only marginally (of the order of ~10%) for the investigated levels of seismic hazard, thus, implying the potential for application of the spectrally equivalent ground-motion records, for seismic fragility and risk assessment at the near-fault sites.

• Structural Engineering and Mechanics
• /
• v.25 no.3
• /
• pp.291-310
• /
• 2007

Current seismic design guidelines in Japan are diverse in the seismic ground strain estimates, because the concepts on a horizontally propagating wave model are not consistent in various seismic design guidelines including gas, water and other underground structures. The purpose of this study is (a) to derive the analytical methods to estimate the ground strains for incident seismic waves, (b) to develop a statistical estimation technique of the ground strains, and finally (c) to compare the theoretical estimation with the observed data which was measured at 441 sites in the 1999 Chi-Chi Earthquake in Taiwan.

• Structural Engineering and Mechanics
• /
• v.88 no.2
• /
• pp.117-128
• /
• 2023

On February 6, 2023, Türkiye woke up with a strong ground motion felt in a wide geography. As a result of the Kahramanmaraş, Pazarcık and Elbistan earthquakes, which took place 9 hours apart, there was great destruction and loss of life. The 2023 Kahramanmaraş earthquakes occurred on active faults known to pose a high seismic hazard, but their effects were devastating. Seismic code spectra were investigated in Hatay, Adıyaman and Kahramanmaraş where destruction is high. The study mainly focuses on the investigation of ground motion parameters of 6 February Kahramanmaraş earthquakes and the correlation between ground motion parameters. In addition, earthquakes greater than Mw 5.0 that occurred in Türkiye were compared with certain seismic parameters. As in the strong ground motion studies, seismic energy parameters such as Arias intensity, characteristic intensity, cumulative absolute velocity and specific energy density were determined, especially considering the duration content of the earthquake. Based on the study, it was concluded that the structures were overloaded far beyond their normal design levels. This, coupled with significant vertical seismic components, is a contributing factor to the collapse of many buildings in the area. In the evaluation made on Arias intensity, much more energy (approximately ten times) emerged in Kahramanmaraş earthquakes compared to other Türkiye earthquakes. No good correlation was found between moment magnitude and peak ground accelerations, peak ground velocities, Arias intensities and ground motion durations in Türkiye earthquakes. Both high seismic components and long ground motion durations caused intense energy to be transferred to the structures. No strong correlation was found between ground motion durations and other seismic parameters. There is a strong positive correlation between PGA and seismic energy parameter AI. Kahramanmaraş earthquakes revealed that changes should be made in the Turkish seismic code to predict higher spectral acceleration values, especially in earthquake-prone regions in Türkiye.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2002.03a
• /
• pp.767-772
• /
• 2002

Free-field ground motion during earthquake is significantly affected by the local site conditions and it is essential in the seismic design to perform the accurate site-specific ground response analysis. In this paper, one-dimensional seismic characteristics of waste landfill are studied based on the vertical propagation of horizontal shear waves through the column of soil/waste. Seismic response analysis is peformed for short-period, long-period and artificial earthquake ground motions using a computer program for seismic response analysis of horizontally layered soil deposits. The computed peak ground accelerations are compared with the values calculated according to Korean seismic design guidelines. The analysis result shows that the long-period earthquake causes the largest peak ground acceleration while the artificial earthquake results in the smallest one.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.28 no.5
• /
• pp.285-294
• /
• 2024

Seismic fragility curves present the conditional probability of damage to target structures due to external seismic load and are widely used in various ways. When constructing such a seismic fragility curve, it is essential to consider various types and numbers of ground motions. In general, the earthquake occurrence characteristics of an area where the target structure of the seismic fragility curve exists are analyzed, and based on this, appropriate ground motions are selected to derive the seismic fragility curve. If the number of selected ground motions is large, the diversity of ground motions is considered, but a large amount of computational time is required. Conversely, if the number of ground motions is too small, the diversity of ground motions cannot be considered, which may distort the seismic fragility curve. Therefore, this study analyzed the relationship between the number of ground motions considered when deriving the seismic fragility curve and the parameters of the seismic fragility curve. Using two example structures, numerical analysis was performed by selecting a random number of ground motions from a total of two hundred, and a seismic fragility curve was derived based on the results. Analysis of the relationship of the parameter of the seismic fragility curve and the number of selected ground motions was performed. As the number of ground motions considered increases, uncertainty in ground motion selection decreases, and when deriving seismic fragility curves considering the same number of ground motions, uncertainty increases relatively as the degree of freedom of the target structure increases. However, considering a relatively large number of ground motions, uncertainty appeared insignificant regardless of increased degrees of freedom. Finally, it is possible that the increase in the number of ground motions could lower the epistemic uncertainty and thus improve the reliability of the results.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.20 no.1
• /
• pp.63-70
• /
• 2016

This study is the sequel of a companion paper (I. Algorithm) for assessment of the seismic performance evaluation of structure using ground motions selected by the proposed algorithm. To evaluate the effect of the correlation structures of selected ground motions on the seismic responses of a structure, three sets of ground motions are selected with and without consideration of the correlation structure. Nonlinear response history analyses of a 20-story reinforced concrete frame are conducted using the three sets of ground motions. This study shows that the seismic responses of the frames vary according to ground motion selection and correlation structures.

• Structural Engineering and Mechanics
• /
• v.53 no.5
• /
• pp.957-979
• /
• 2015

The present paper investigates the influence of the orientation of the ground-motion reference axes, the seismic incident angle and the seismic intensity level on the inelastic response of asymmetric reinforced concrete buildings. A single storey asymmetric building is analyzed by nonlinear dynamic analyses under twenty bi-directional ground motions. The analyses are performed for many angles of incidence and four seismic intensity levels. Moreover three different pairs of the horizontal accelerograms corresponding to the input seismic motion are considered: a) the recorded accelerograms, b) the corresponding uncorrelated accelerograms, and c) the completely correlated accelerograms. The nonlinear response is evaluated by the overall structural damage index. The results of this study demonstrate that the inelastic seismic response depends on the orientation of the ground-motion reference axes, since the three individual pairs of accelerograms corresponding to the same ground motion (recorded, uncorrelated and completely correlated) can cause different structural damage level for the same incident angle. Furthermore, the use of the recorded accelerograms as seismic input does not always lead to the critical case of study. It is also shown that there is not a particular seismic incident angle or range of angles that leads to the maximum values of damage index regardless of the seismic intensity level or the ground-motion reference axes.