• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.592-599
• /
• 2020

This paper proposes a Convolutional Recurrent Neural Net (CRNN) structure that can simultaneously reflect both static and dynamic characteristics of seismic waveforms for various earthquake events classification. Addressing various earthquake events, including not only micro-earthquakes and artificial-earthquakes but also macro-earthquakes, requires both effective feature extraction and a classifier that can discriminate seismic waveform under noisy environment. First, we extract the static characteristics of seismic waveform through an attention-based convolution layer. Then, the extracted feature-map is sequentially injected as input to a multi-input single-output Long Short-Term Memory (LSTM) network structure to extract the dynamic characteristic for various seismic event classifications. Subsequently, we perform earthquake events classification through two fully connected layers and softmax function. Representative experimental results using domestic and foreign earthquake database show that the proposed model provides an effective structure for various earthquake events classification.

• Earthquakes and Structures
• /
• v.23 no.3
• /
• pp.315-328
• /
• 2022

Structures designed for wind have an opposite design approach to those designed for earthquakes. These structures are usually reliable if they are constructed in an area where there is almost no or less severe earthquake. However, as seismic activity is unpredictable and it can occur anytime and anywhere, the seismic safety of structures designed for wind must be assessed. Moreover, the design approaches of wind and earthquake systems are opposite where wind design considers higher stiffness but earthquake designs demand a more flexible structure. For this reason, a novel Machine learning framework is proposed that is used to assess and classify the seismic safety of the structures designed for wind load. Moreover, suitable criteria is defined for the design of wind resistance structures considering seismic behavior. Furthermore, the structural behavior as a result of dynamic interaction between superstructure and substructure during seismic events is also studied. The proposed framework achieved an accuracy of more than 90% for classification and prediction as well, when applied to new structures and unknown ground motions.

• Journal of the Korean Geotechnical Society
• /
• v.35 no.11
• /
• pp.121-129
• /
• 2019

An Earthquake Early Warning (EEW) system is a technology that alerts people to an incoming earthquake by using P waves that are detected before the arrival of more severe seismic waves. P-wave analysis is therefore an important factor in the production of rapid seismic information as it can be used to quickly estimate the earthquake magnitude and epicenter through the amplitude and predominant period of the observed P-wave. However, when a large-magnitude teleseismic earthquake is observed in a local seismic network, the significantly attenuated P wave phases may be mischaracterized as belonging to a small-magnitude local earthquake in the initial analysis stage. Such a misanalysis may be sent to the public as a false alert, reducing the credibility of the EEW system and potentially causing economic losses for infrastructure and industrial facilities. Therefore, it is necessary to develop methods that reduce misanalysis. In this study, the possibility of seismic misclassifying teleseimic earthquakes as local events was reviewed using the Filter Bank method, which uses the attenuation characteristics of P waves to classify local and outside Korean peninsula (regional and teleseismic) events with filtered waveform depending on frequency and epicenter distance. The data used in our analysis were analyzed for maximum P_v values using 463 events with local magnitudes (2 < M_L ≦ 3), 44 (3 < M_L ≦ 4), 4 (4 < M_L ≦ 5), 3 (M_L > 5), and 89 outside Korean peninsula earthquakes recorded by the KMA seismic network. The results show that local and telesesimic earthquakes can be classified more accurately when combination of filtering bands of No. 3 (6-12 Hz) and No. 6 (0.75-1.5 Hz) is applied.

• 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.

• Earthquakes and Structures
• /
• v.9 no.5
• /
• pp.983-997
• /
• 2015

This research evaluates the soil conditions for seismic stations situated in Romania using the horizontal-to-vertical spectral ratio (HVSR). The strong ground motion database assembled for this study consists of 179 analogue and digital strong ground motion recordings from four intermediate-depth Vrancea seismic events with $M_w{\geq}6.0$. In the first step of the analysis, the influence of the earthquake magnitude and source-to-site distance on the H/V curves is evaluated. Significant influences from both the earthquake magnitude and hypocentral distance are found especially for soil class A sites. Next, a site classification method proposed in the literature is applied for each seismic station and the soil classes are compared with those obtained from borehole data and from the topographic slope method. In addition, the success and error rates of this method are computed and compared with other studies from the literature. A more in-depth analysis of the H/V results is performed using data from seismic stations in Bucharest and a comparison of the free-field and borehole H/V curves is done for three seismic stations. The results show large differences between the free-field and the borehole curves. As a conclusion, the results from this study represent an intermediary step in the evaluation of the soil conditions for seismic stations in Romania and the need to perform more detailed soil classification analysis is highly emphasized.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.17 no.4
• /
• pp.10-27
• /
• 2014

HAZUS-MH is a GIS-based computer program that estimates potential losses from multi-hazard phenomena: earthquakes, floods and hurricanes. With respect to seismic disaster, characteristics of a hypothetical or actual earthquake are entered into HAZUS. Then HAZUS estimates the intensity of ground shaking and calculates the correspondent losses. In this study, HAZUS was used as a part of the preparations of the future seismic events at a coastal plant facility area. To reliably characterize the target facility area, many geotechnical characteristics data were synthesized from the existing site investigation reports. And the buildings and facilities were sorted by analyzing their material and structural characteristics. In particular, the study area was divided into 17 blocks taking into account the situation of both land development and facility distribution. The ground conditions of blocks were categorized according to the site classification scheme for earthquake-resistant design. Moreover, seismic fragility curves of a main facilities were derived based on the numerical modeling and were incorporated into the database in HAZUS. The results estimated in the study area using HAZUS showed various seismic damage and loss potentials depending on site conditions and structural categories. This case study verified the usefulness of the HAZUS for estimating earthquake losses in coastal facility areas.

• 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.

• Journal of the Korean Geotechnical Society
• /
• v.28 no.3
• /
• pp.35-43
• /
• 2012

Following the companion paper (I. Comparisons with Well-known Seismic Code and Site Response Characteristics), several acceleration data recorded during recent earthquake events in Korea were analyzed to verify the suitability of the proposed two-parameters site classification system and the corresponding site coefficients. For all of rock-soil site pairs less than 30 km distant, response spectrums and corresponding site coefficients, $F_a$ and $F_v$, were determined. Unfortunately, some of data have an eccentric error, where the spectral acceleration of rock site is more amplified than that of soil site. The $F_a$ and $F_v$ for all of pairs except the pairs of error were compared with those in the current code and the proposed system. The $F_a$ and $F_v$ from the recorded motions show definitely different trend from that of the current code. In addition, the site coefficients from recorded motions at four 765 kV substation sites, which are several hundred meters distant, have a remarkably similar trend and absolute values to those in proposed two-parameters site classification system. Based on earthquake motions recorded in domestic areas including data from the four 765 kV substation sites, the two-parameters site classification and site coefficients are superior to the results obtained from the current Korean seismic code.

• Journal of Science and Technology Studies
• /
• v.19 no.3
• /
• pp.51-117
• /
• 2019

On 15th November 2017, the coastal city of Pohang, located in the Southeastern part of South Korea was shaken by a magnitude 5.4 earthquake. The earthquake displaced more than 1,700 residents and caused more than $ 300 million dollars of economic loss. It was the second most damaging earthquake in the history of Korea. Soon after the earthquake, a group of scientists raised a possible link between the first Enhanced Geothermal System (EGS) project and the earthquake. At the same time, another group of scientists put forward a different hypothesis of the causation of the earthquake claiming that it was caused by the geological movements that were initiated by the Great Tohoku Earthquake in 2011. Since then, there were scientific debates between the two different groups of scientists. The scientific debate on the causation of the earthquake has been concluded temporarily by the Research Investigatory Committee on the Pohang Earthquake in 2019. The research committee concluded that the earthquake was caused by the Pohang EGS system: this means that the earthquake can be defined not as a natural earthquake, but as an artificially triggered earthquake. This article is to examine the Pohang earthquake can be defined as an Anthropocenic event. The newly suggested concept, the Anthropocene is a relatively novel term to classify the earthly strata and their relationship to geological time. The current geological period should be defined by human activities and man-made earthly environment. Although the term is basically related to geological classification, the Anthropocene has been widely debated amongst humanist and social science scholars. The current disastrous situation of our planet also implies with the Anthropocene. This paper is to discuss how to understand anthropogenic events. In particular, the paper pays attention to two different scholarly positions on the Anthropocene: Isabelle Stenger's Gaia theory and Barbara Herrnstein Smith's relativist theory. The former focuses on the earthly inevitable catastrophe of Anthropocene while the latter suggests to situate and contextualise anthropogenic events. On the basis of the theoretical positions, the article is to analyse how the Pohang earthquake can be located and situated.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.61 no.3
• /
• pp.101-112
• /
• 2019

This study developed selection criteria of small-scales reservoirs, having under $300,000m^3$ storage capacity, for the Emergency Action Plan(EAP) establishment in order to reduce the disaster risks of the reservoir's failures. Those reservoirs are out of ranges of Korean EAP establishment standard, but have potential risk of disasters as they have often failed by the recent extreme rainfall events and earthquakes, causing economical and life losses. The problem of reservoir aging is also one of the reasons of them. In this study, the developed selection criteria of small reservoirs for EAP establishment are storage capacity, embankment height, reservoir age, heavy rain factor and earthquake factor. These criteria were selected based on the review of the existing EAP establishment guidelines, analysis of the past dam failure cases, and the previous related studies. The quantification of these criteria were conducted for the practical applications in the fields, and applied to 67 previous failures in order to investigate the relation of each criteria with these failures. The earthquake factor found to be the highest relations followed by heavy rain factors, combination of earthquake and heavy rain factors, and reservoir age. The classification was made as observation and review groups for EAP establishments based on overlapping numbers of each criteria. This classifications applied to 354 reservoirs designated as having the potential disaster risk by MOIS, and showed 38.4% of observation and 11.9% of review groups. Anticipatory monitoring and regular inspection should be made by professional facility managers for the observation group, and necessity of EAP establishment should be assessed for the review group based on the downstream status and financial budget.