• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.169-169
• /
• 2020

According to the IPCC analysis, severe climate changes are projected to occur in Korea as the temperature is expected to rise by 3.2 ℃, the precipitation by 15.6% and the sea level by 27cm by 2050. It is predicted that the occurrence of abnormal climate phenomena - especially those such as increase of concentrated precipitation and extreme heat in the summer season and severe drought in the winter season - that have happened in Korea in the past 30 years (1981-2010) will continuously be intensified and accelerated. As a result, the impact on and vulnerability of the water management sector is expected to be exacerbated. This research aims to predict the climate change impacts on streamflow of Daecheong Lake area of Geum River in South Korea during the summer and winter seasons, which show extreme meteorological events, and ultimately develop an integrated policy model in response. We projected and compared the streamflow changes of Daecheong Lake area of Geum River in South Korea in the near future period (2020-2040) and the far future period (2041-2060) with the reference period (1991-2010) using the HEC-HMS model. The data from a global climate model HadGEM2-AO, which is the fully-coupled atmosphere-ocean version of the Hadley Centre Global Environment Model 2, and RCP scenarios (RCP4.5 and RCP8.5) were used as inputs for the HEC-HMS model to identify the river basins where cases of extreme flooding or drought are likely to occur in the near and far future. The projections were made for the summer season (July-September) and the winter season(November-January) in order to reflect the summer monsoon and the dry winter. The results are anticipated to be used by policy makers for preparation of adaptation plans to secure water resources in the nation.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.1
• /
• pp.51-59
• /
• 2021

The deep learning model can produce false prediction results due to inputs that deviate from training data through variation, which leads to fatal accidents in areas such as autonomous driving and security. To ensure reliability of the model, the model's coping ability for exceptional situations should be verified through various mutations. However, previous studies were carried out on limited scope of models and used several mutation types without separating them. Based on the CIFAR10 data set, widely used dataset for deep learning verification, this study carries out reliability verification for total of six models including various commercialized models and their additional versions. To this end, six types of input mutation algorithms that may occur in real life are applied individually with their various parameters to the dataset to compare the accuracy of the models for each of them to rigorously identify vulnerabilities of the models associated with a particular mutation type.

• Journal of Industrial Convergence
• /
• v.21 no.1
• /
• pp.159-165
• /
• 2023

While the frequency of seismic occurrence has been increasing recently, the domestic seismic response system is weak, the objective of this research is to compare and analyze the seismic vulnerability of buildings using statistical analysis and machine learning techniques. As the result of using statistical technique, the prediction accuracy of the developed model through the optimal scaling method showed about 87%. As the result of using machine learning technique, because the accuracy of Random Forest method is 94% in case of Train Set, 76.7% in case of Test Set, which is the highest accuracy among the 4 analyzed methods, Random Forest method was finally chosen. Therefore, Random Forest method was derived as the final machine learning technique. Accordingly, the statistical analysis technique showed higher accuracy of about 87%, whereas the machine learning technique showed the accuracy of about 76.7%. As the final result, among the 22,296 analyzed building data, the seismic vulnerabilities of 1,627(0.1%) buildings are expected as more dangerous when the statistical analysis technique is used, 10,146(49%) buildings showed the same rate, and the remaining 10,523(50%) buildings are expected as more dangerous when the machine learning technique is used. As the comparison of the results of using advanced machine learning techniques in addition to the existing statistical analysis techniques, in spatial analysis decisions, it is hoped that this research results help to prepare more reliable seismic countermeasures.

• Journal of Korean Society of Forest Science
• /
• v.112 no.3
• /
• pp.340-351
• /
• 2023

Landslides are major natural geological hazards that cause enormous property damage and human casualties annually. The vulnerability of mountainous areas to landslides is further exacerbated by the impacts of climate change. Soil depth is a crucial parameter in landslide and debris flow analysis, and plays an important role in the evaluation of watershed hydrological processes that affect slope stability. An accurate method of estimating soil depth is to directly investigate the soil strata in the field. However, this requires significant amounts of time and money; thus, numerous models for predicting soil depth have been proposed. However, they still have limitations in terms of practicality and accuracy. In this study, 71 seismic survey results were collected from domestic mountainous areas to estimate soil depth on hill slopes. Soil depth was estimated on the basis of a shear wave velocity of 700 m/s, and a database was established for slope angle, elevation, and soil depth. Consequently, the statistical characteristics of soil depth were analyzed, and the correlations between slope angle and soil depth, and between elevation and soil depth were investigated. Moreover, various soil depth prediction models based on slope angle were investigated, and corrected linear and exponential soil depth prediction models were proposed.

• Journal of Korea Water Resources Association
• /
• v.55 no.7
• /
• pp.481-494
• /
• 2022

This study proposed a evaluation of the monthly vulnerable period for groundwater level management in the Miho stream watershed and a technique for evaluating the vulnerable period for future groundwater level management using LSTM. Observation data from groundwater level and precipitation observation stations in the Miho stream watershed were collected, LSTM was constructed, predicted values for precipitation and groundwater levels from 2020 to 2022 were calculated, and future groundwater management was evaluated when vulnerable. In order to evaluate the vulnerable period of groundwater level management, the correlation between groundwater level and precipitation was considered, and weights were calculated to consider changes caused by climate change. As a result of the evaluation, the Miho stream watershed showed high vulnerability to underground water management in February, March, and June, and especially near the Cheonan Susin observation well, the vulnerability index for groundwater level management is expected to deteriorate in the future. The results of this study are expected to contribute to the evaluation of the vulnerable period of groundwater level management and the derivation of preemptive countermeasures to the problem of groundwater resources in the basin by presenting future prediction techniques using LSTM.

• Spatial Information Research
• /
• v.20 no.1
• /
• pp.9-18
• /
• 2012

South Korea has many landslides caused by heavy rains during summer time recently and the landslides continue to cause damages in many places. These landslides occur repeatedly each year, and the frequency of landslides is expected to increase in the coming future due to dramatic global climate change. In Korea, 81.5% of the population is living in urban areas and about 1,055 million people are living in Seoul. In 2011, the landslide that occurred in Seocho-dong killed 18 people and about 9% of Seoul's area is under the same land conditions as Seocho-dong. Even the size of landslide occurred in a city is small, but it is more likely to cause a big disaster because of a greater population density in the city. So far, the effort has been made to identify landslide vulnerability and causes, but now, the new dem and arises for the prediction study about the areal extent of disaster area in case of landslides. In this study, the diffusion model of the landslide disaster area was established based on Cellular Automata(CA) to analyze the physical diffusion forms of landslide. This study compared the accuracy with the Seocho-dong landslide case, which occurred in July 2011, applying the SCIDDICA model and the CAESAR model. The SCIDDICA model involves the following variables, such as the movement rules and the topographical obstacles, and the CAESAR model is also applied to this process to simulate the changes of deposition and erosion.

• Journal of the Korea Society for Simulation
• /
• v.30 no.2
• /
• pp.49-64
• /
• 2021

M&S techniques are widely used as scientific means to systematically develop response plans to chemical and biological (CB) hazards. However, while the theoretical area of hazard dispersion modeling has achieved remarkable practical results, the operational analysis area to simulate CB hazard response plans is still in an early stage. This paper presents a model to simulate CB hazard response plans such as detection, protection, and decontamination. First, we present a possible way to display high-fidelity hazard dispersion in a combat simulation model, taking into account weather and terrain conditions. We then develop an improved vulnerability model of the combat simulation model, in order to simulate CB damage of combat simulation entities based on other casualty prediction techniques. In addition, we implement tactical behavior task models that simulate CB hazard response plans such as detection, reconnaissance, protection, and decontamination. Finally, we explore its feasibility by analyzing contamination detection effects by distributed CB detectors and decontamination effects according to the size of the {contaminated, decontamination} unit. We expect that the proposed model will be partially utilized in disaster prevention and simulation training area as well as analysis of combat effectiveness analysis of CB protection system and its operational concepts in the military area.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.1
• /
• pp.225-234
• /
• 2019

Bitcoin is a cryptocurrency with characteristics such as de-centralization and distributed ledger, and these features are maintained through a mining system called "proof of work". In the mining system, mining difficulty is adjusted to keep the block generation time constant. However, Bitcoin's current method to update mining difficulty does not reflect the future hash power, so the block generation time can not be kept constant and the error occurs between designed time and real time. This increases the inconsistency between block generation and real world and causes problems such as not meeting deadlines of transaction and exposing the vulnerability to coin-hopping attack. Previous studies to keep the block generation time constant still have the error. In this paper, we propose a machine-learning based method to reduce the error. By training with the previous hash power, we predict the future hash power and adjust the mining difficulty. Our experimental result shows that the error rate can be reduced by about 36% compared with the current method.

• Advances in concrete construction
• /
• v.14 no.3
• /
• pp.153-167
• /
• 2022

The interaction between blast load and structures, as well as the interaction among structural members may well affect the structural response and damages. Therefore, it is necessary to analyse more realistic reinforced concrete structures in order to gain an extensive knowledge on the possible structural response under blast load effect. Among all the civilian structures, columns are considered to be the most vulnerable to terrorist threat and hence detailed investigation in the dynamic response of these structures is essential. Therefore, current research examines the effect of blast loads on the reinforced concrete columns via development of Pressure- Impulse (P-I) diagrams. In the finite element analysis, the level of damage on each of the aforementioned RC column will be assessed and the response of the RC columns when subjected to explosive loads will also be identified. Numerical models carried out using LS-DYNA were compared with experimental results. It was shown that the model yields a reliable prediction of damage on all RC columns. Validation study is conducted based on the experimental test to investigate the accuracy of finite element models to represent the behaviour of the models. The blast load application in the current research is determined based on the Lagrangian approach. To develop the designated P-I curves, damage assessment criteria are used based on the residual capacity of column. Intensive investigations are implemented to assess the effect of column dimension, concrete and steel properties and reinforcement ratio on the P-I diagram of RC columns. The produced P-I models can be applied by designers to predict the damage of new columns and to assess existing columns subjected to different blast load conditions.

• Korean Journal of Remote Sensing
• /
• v.34 no.3
• /
• pp.545-552
• /
• 2018

Grounding line is used as evidence of the mass balance showing the vulnerability of Antarctic glaciers and ice shelves. In this research, we utilized a high resolution digital elevation model of glacier surface derived by recently launched satellites to estimate the position of grounding line of Campbell Glacier in East Antarctica. TanDEM-X and TerraSAR-X data in single-pass interferometry mode were acquired on June 21, 2013 and September 10, 2016 and CryoSat-2 radar altimeter data were acquired within 15 days from the acquisition date of TanDEM-X. The datasets were combined to generate a high resolution digital elevation model which was used to estimate the grounding line position. During the 3 years of observation, there weren't any significant changes in grounding line position. Since the average density of ice used in estimating grounding line is not accurately known, the variations of the grounding line was analyzed with respect to the density of ice. There was a spatial difference from the grounding line estimated by DDInSAR whereas the estimated grounding line using the characteristics of the surface of the optical satellite images agreed well when the ice column density was about $880kg/m^3$. Although the reliability of the results depends on the vertical accuracy of the bathymetry in this study, the hydrostatic ice thickness has greater influence on the grounding line estimation.