In this paper, comparative analysis of the 9.12 Gyeongju and 11.15 Pohang earthquakes was conducted in order to provide probable explanations and reasons for the damage observed in the 11.15 Pohang earthquake from both earthquake and structural engineering perspectives. The damage potentials like Arias intensity, effective peak ground acceleration, etc observed in the 11.15 Pohang earthquake were generally weaker than those of the 9.12 Gyeongju earthquake. However, in contrast to the high-frequency dominant nature of the 9.12 Gyeongju earthquake records, the spectral power of PHA2 record observed in the soft soil site was highly concentrated around 2Hz. The base shear around 2 Hz frequency was as high as 40% building weight. This frequency band is very close to the fundamental frequency of the piloti-type buildings severely damaged in the northern part of Pohang. Unfortunately, in addition to inherent vertical irregularity, most of the damaged piloti-type buildings had plan irregularity as well and were non-seismic. All these contributed to the fatal damage. Inelastic dynamic analysis indicated that PHA2 record demands system ductility capacity of 3.5 for a structure with a fundamental period of 0.5 sec and yield base shear strength of 10% building weight. The system ductility level of 3.5 seems very difficult to be achievable in non-seismic brittle piloti-type buildings. The soil profile of the PHA2 site was inversely estimated based on deconvolution technique and trial-error procedure with utilizing available records measured at several rock sites during the 11.15 Pohang earthquake. The soil profile estimated was very typical of soil class D, implying significant soil amplification in the 11.15 Pohang earthquake. The 11.15 Pohang earthquake gave us the expensive lesson that near-collapse damage to irregular and brittle buildings is highly possible when soil is soft and epicenter is close, although the earthquake magnitude is just minor to moderate (M 5+).

This paper investigates seismic damage potential of recent September 12 M5.8 Gyeongju earthquake from diverse earthquake engineering perspectives using the accelerograms recorded at three stations near the epicenter. In time domain, strong motion durations are evaluated based on the accelerograms and compared with statistical averages of the ground motions with similar magnitude, epicentral distance and soil conditions, while Fourier analysis using FFT is performed to identify damaging frequency contents contained in the earthquake. Effective peak ground accelerations are evaluated from the calculated response spectra and compared with apparent peak ground accelerations and the design spectrum in KBC 2016. All these results are used to consistently explain the reason why most of seismic damage in the earthquake was concentrated on low-rise stiff buildings but not quite significant. In order to comparatively appraise the damage potential, the constant ductility spectrum constructed from the Gyeongju earthquake is compared with that of the well-known 1940 El Centro earthquake. Deconvolution analysis by using one accelerogram speculated to be recorded at a stiff soil site is also performed to estimate the soil profile conforming to the response spectrum characteristics. Finally, response history analysis for 39- and 61-story tall buildings is performed as a case study to explain significant building vibration felt on the upper floors of some tall buildings in Busan area during the Gyeongju earthquake. Seismic design and retrofit implications of M5.8 Gyeongju earthquake are summarized for further research efforts and improvements of relevant practice.

In this paper, firstly, acceleration-time histories were generated by varying strong motion duration in the frequency domain for application to a seismically isolated nuclear power structure, so as to examine the effects of strong motion duration on the behavior of the structure. Secondly, real recorded earthquakes were modified to match the target response spectrum based on the revised SRP 3.7.1(2007) and the modified time histories were applied to the analysis of a seismically isolated nuclear power structure. The obtained values of acceleration and displacement responses of the structure were, finally, compared with the values obtained in case of applying acceleration-time histories generated in the frequency domain to the structure.

이 논문은 원전구조물의 내진설계에 적용되는 인공지진파의 강진지속시간과 포락함수에 대한 현행 국내 설계기준의 개선과 보완을 위해서 필요한 기반연구에 관한 내용을 다루고 있다. USNRC와 ASCE 4-98에서 제안한 응답스펙트럼과 강진지속시간에 대한 규정이 현재 통상적으로 사용되고 있으며, 첫 번째로 두 기준에 대한 비교와 검토를 수행하였다. 다음으로 총 209개의 암반사이트에서 실제 계측된 규모 5.0 이상인 강진기록을 ASCE 4-98의 강진지속시간기준에 적용한 결과를 통계 처리하여 지진규모에 대한 함수로 표현되는 강진지속시간의 실험적 예측모델을 제시하였다. 마지막으로 강진지속시간이 원전구조물의 지진응답특성에 미치는 영향을 파악하기 위하여 6초에서 20초까지 약 2초 간격으로 강진지속시간을 달리하는 10가지 Case에 대한 인공지진파를 각 30개씩 작성하고, 이들을 적용하여 대만 Hualien 지진시험구조물과 국내 울진 원자력발전소 원자로 격납구조물에 대한 광범위한 지진응답해석을 수행하고 그 결과를 분석하였다.