• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.130-138
• /
• 2021

This paper presents an effective structure by applying various normalization to Convolutional Neural Networks (CNN) for seismic event classification. Normalization techniques can not only improve the learning speed of neural networks, but also show robustness to noise. In this paper, we analyze the effect of input data normalization and hidden layer normalization on the deep learning model for seismic event classification. In addition an effective model is derived through various experiments according to the structure of the applied hidden layer. As a result of various experiments, the model that applied input data normalization and weight normalization to the first hidden layer showed the most stable performance improvement.

• Wind and Structures
• /
• v.32 no.4
• /
• pp.283-292
• /
• 2021

Performance-based seismic design (PBSD) is currently used for retrofitting of older buildings and the design of new buildings. Whereas, application of performance-based design for wind load is still under development. The tendency has been in the codes to increase wind hazard based on recent recorded events. Since tall buildings are highly susceptible to wind load, necessity for developing a framework for performance-based wind design (PBWD) has intensified. Only a few guidelines such as ASCE (2019) provide information on using PBWD as an alternative for code prescriptive wind design. Though wind hazards, performance objectives, analysis techniques, and acceptance criteria are explained, no recommendations are provided for several aspects like how to select a proper level of wind hazard for each target performance criterion. This paper is an attempt to explain current design philosophy for wind and seismic loads and inherent connection between the components of PBSD for development of a framework for PBWD of tall buildings. Recognizing this connection, a framework for PBWD based on limits set for serviceability and strength is also proposed. Also, the potential for carrying out PBWD in line with ASCE 7-16 is investigated and proposed in this paper.

• Earthquakes and Structures
• /
• v.22 no.1
• /
• pp.15-23
• /
• 2022

One of the key tools in assessing the seismic vulnerability of the structures is the use of fragile functions, which is the possibility of damage from a particular damage surface for several levels of risk from the seismic movements of the earth. The aim of this study is to investigate the effect of two categories of earthquake events on the fragile curve (FRC) of the steel construction system. In this study, the relative lateral displacement of the structures is considered as a damage criterion. The limits set for modifying the relative lateral position in the HAZUS instruction are used to determine the failure modes, which include: slight, moderate, extensive and complete. The results show, as time strong-motion increases, the probability of exceeding (PoE) increases (for Peak ground acceleration (PGA) less than 0.5). The increase in seismic demand increases the probability of exceeding. In other words, it increases the probability of exceeding, if the maximum earthquake acceleration increases. Also, 7-storey model in extensive mode has 20 and 26.5% PoE larger than 5- and 3-storey models, respectively.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.5
• /
• pp.217-220
• /
• 2024

Predicting earthquake occurrences accurately is challenging, and preparing all buildings with seismic design for such random events is a difficult task. Analyzing building features to predict potential damage and reinforcing vulnerabilities based on this analysis can minimize damages even in buildings without seismic design. Therefore, research analyzing the efficiency of building damage prediction models is essential. In this paper, we compare the accuracy of earthquake damage prediction models using machine learning classification algorithms, including Random Forest, Extreme Gradient Boosting, LightGBM, and CatBoost, utilizing data from buildings damaged during the 2015 Nepal earthquake.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 1997.04a
• /
• pp.41-48
• /
• 1997

In seismology, instrumental data is covering from the end of the last century and showing large uncertainties in earthquake parameters before 1960. The number and quality of seismological stations have been in steady increased over all the past decades of this century, and this development is still going on. In Korea, reliable instrumental data is only available since 1978 from KMA(Korea Meteorological Administration). However instrumental earthquake observation have started since 1905 by Japanese and seven seismic stations were in operation in 1941. We have compiled and analyzed the early instrumental earthquake data between 1905 and 1942. Total 533 events were analyzed and for 60 events, their epicentral coordinates and magnitude were determined

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2003.09a
• /
• pp.3-7
• /
• 2003

The velocity structure beneath the CHNB broadband station is determined by receiver function analysis using by from teleseismic P waveforms. The detailed broadband receiver functions are obtained by stacking method for source-equalized vertical, radial and tangential components of teleseismic P waveforms. A time domain inversion uses the stacked radial receiver function to determine vertical P wave velocity structure beneath the station. The crustal velocity structures beneath the stations are estimated using the receiver function inversion method in the case at the crustal model parameterized by many thin, flat-lying, homogeneous layers. Events divide into 4 groups. four azimuths corresponding to events in group a(southwest), b(south), c(southeast), d(northeast). The result of crust at model inversion shows the crustal velocity structure beneath the CHNB station varies smoothly with increasing depth. The conard discontinuity lies around 18 km and moho discontinuity lies range from 30 to 34 km.

• Structural Engineering and Mechanics
• /
• v.74 no.6
• /
• pp.789-807
• /
• 2020

Multiple earthquakes that occur during short seismic intervals affect the inelastic behavior of the structures. Sequential ground motions against the single earthquake event cause the building structure to face loss in stiffness and its strength. Although, numerous research studies had been conducted in this research area but still significant limitations exist such as: 1) use of traditional design procedure which usually considers single seismic excitation; 2) selecting a seismic excitation data based on earthquake events occurred at another place and time. Therefore, it is important to study the effects of successive ground motions on the framed structures. The objective of this study is to overcome the aforementioned limitations through testing a two storey RC building structural model scaled down to 1/10 ratio through a similitude relation. The scaled model is examined using a shaking table. Thereafter, the experimental model results are validated with simulated results using ETABS software. The test framed specimen is subjected to sequential five artificial and four real-time earthquake motions. Dynamic response history analysis has been conducted to investigate the i) observed response and crack pattern; ii) maximum displacement; iii) residual displacement; iv) Interstorey drift ratio and damage limitation. The results of the study conclude that the low-rise building model has ability to resist successive artificial ground motion from its strength. Sequential artificial ground motions cause the framed structure to displace each storey twice in correlation with vary first artificial seismic vibration. The displacement parameters showed that real-time successive ground motions have a limited impact on the low-rise reinforced concrete model. The finding shows that traditional seismic design EC8 requires to reconsider the traditional design procedure.

• Earthquakes and Structures
• /
• v.12 no.1
• /
• pp.13-21
• /
• 2017

Recent seismic events occurred in Italy (Emilia-Romagna 2012, Abruzzo 2009) and worldwide (New Zealand 2010 and 2011) highlighted some of the weaknesses of precast concrete industrial buildings, especially those related to the connecting systems traditionally employed to fasten the cladding panels to the internal framing. In fact, one of the most commons fails it is possible to observe in such structural typologies is related to the out-of-plane collapse of the external walls due to the unsatisfactory behaviour of the connectors used to join the panels to the perimeter beams. In this work, the strengthening of a traditional industrial building, assumed as a case study, made by precast reinforced concrete is proposed by the adoption of a dual system allowing the reinforcement of the structure by acting both internally; by pendular columns and, externally, on the walls. In particular, traditional connections at the top of the walls are substituted by devices able to work as a slider with vertical axis while, the bottom of the walls is equipped with two or more hysteretic dampers working on the uplift of the cladding panels occurring under seismic actions. By means of this approach, the structure is stiffened; obtaining a reduction of the lateral drifts under serviceability limit states. In addition, its seismic behaviour is improved due to the additional source of energy dissipation represented by the dampers located at the base of the walls. The effectiveness of the suggested retrofitting approach has been checked by comparing the performance of the retrofitted structure with those of the structure unreinforced by means of both pushover and Incremental Dynamic Analyses (IDA) in terms of behaviour factor, assumed as a measure of the ductility capacity of the structure.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.5 no.1 s.16
• /
• pp.23-32
• /
• 2005

A significant development has been made on earthquake resistant designs on many structures since the Korean government has begun modern earthquake hazard reduction programs after recognizing potential disastrous consequences of seismic events following the Kobe earthquake in 1995. However, some structures such as slope structures still haven't get ready for their own seismic design guidelines in Korea. Therefore, only a few organizations of Korea adopt seismic design for slopes relying on designers' judgments at present. This paper introduces domestic and foreign research activities on seismic slope stability and an idea of Korean earthquake resistant design method for slopes including alternatives of earthquake resistant design application according to designers' judgment considering construction budget, importance, restoration and so on. Afterwards, seismic data accumulation on slope stability of Korea is necessary to induce a more definite Korean earthquake resistant design method.

• Journal of the Korean Geotechnical Society
• /
• v.28 no.5
• /
• pp.55-65
• /
• 2012

The purpose of this study is to estimate seismic damage for harbor site considering dynamic amplification characteristics. First of all, a series of ground response analysis is performed and then correlation equations between rock outcrop accelerations and peak ground accelerations (PGAs) are determined. These equations are saved into DB and when an earthquake occurs, PGAs are determined by them as soon as possible. For earthquake events, seismic damage grades of harbor structures are determined by using the correlated PGAs and fragility curves of harbor structures in real time. In this study, seismic damage was estimated and classified into several grades by applying two hypothetical earthquakes.