• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.269-269
• /
• 2023

기후 변화로 인한 강우 패턴의 변화는 도심지 방재성능 목표를 상회하는 홍수로 이어져 침수피해를 가중시키고 있다. 이로 인한 도시침수 피해를 저감하기 위하여 도시침수 예측모형 개발이 활발히 이루어지고 있으나, 대규모 관망으로 이루어진 복잡한 도심지 우수관망을 모의하기 때문에 분석속도가 느려 실시간 예측 적용에 한계점이 있다. 도시침수 분석에 가장 많이 활용되는 대표적인 모형인 SWMM(Storm Water Management Model)은 복잡한 관망을 비교적 빠르고 정확히 해석할 수 있어 유용하지만, 이 또한 대도심의 우수관망 모의 시 많은 시간이 소요되며, 관망 정밀도 기준이 정의되어 있지 않아 분석에 어려움이 있다. 이러한 문제점을 해결하기 위하여 본 연구에서는 관망 간소화 기법(유역면적의 밀도, 관거 직경, 관로의 길이 등)을 적용하고, 이에 따른 주요 지선과 간선의 수위 변화와 침수흔적도를 비교하여 분석결과의 정확성을 담보하는 관망 간소화 수준을 파악하고 도시침수 분석 시 적정 간소화 기준과 자동 간소화 방안을 제시하고자 한다. 도시침수 분석 시 우수관망 자동 간소화를 위하여 Python을 활용한 코드를 작성하였으며, SWMM의 .inp 파일을 읽어들여 Dataframe형태로 저장한 후 분석을 위한 데이터 가공, 간소화 기준에 따른 분류, 간소화 대상 수리·수문인자 연산, 인접 간선에 연결, 간소화된 .inp파일 저장의 총 6단계로 구성하였다. 연구 대상지역은 도림천 유역으로 설정하였으며, 초기자료는 맨홀 30,469, 관거 32,443, 소유역 30,586개로 이루어져 있으며, 모의 시간은 약 2시간 30분이 소요되었다. 유역면적 100x100 미만을 대상으로 수행 시 맨홀 9,965, 관거 10,464, 소유역 9,240개로 관거의 복잡도가 약 1/3 감소하였으며, 모의 시간은 약 43분으로 기존대비 약 72% 단축되는 것으로 나타났다. 실제 침수가 발생한 주요지점들을 비교한 결과 R² 0.85 ~ 0.92로 예측모형의 정확도에 큰 영향을 끼치지 않는 것으로 나타났다. 도시침수모형 최적 간소화를 통해 모형의 복잡성을 줄이고, 계산량을 줄여 모형의 수행시간을 단축시킬 수 있으며, 불필요한 우수관망을 제거하거나 병합함으로써, 모형의 예측력 향상과 분석과 해석에 효율적으로 사용될 수 있을 것으로 기대한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.458-458
• /
• 2021

하천의 환경기능과 생태에 관한 관심이 증가하며 생태를 고려한 하천 유지 유량의 산정이 필요하다. 수문 특성의 물리적 서식처에 관한 연구는 PHABSIM, River2D 등의 소프트웨어 적용을 통한 유지유량증분법(IFIM), 수문 인자별 서식처적합도지수(HSI)를 기반으로 연구되고 있지만, 하천의 수질을 고려한 서식처 연구는 각 수질 인자별 서식처적합도지수 자료의 부족으로 하천유지유량 산정에 반영이 어려운 실정이다. 본 연구는 국내 하천의 수질·생태 모니터링 자료를 바탕으로 수온, DO 등의 수질 인자에 대한 피라미의 서식처적합도지수를 권역 단위로 산정했다. 수질 및 어류조사 자료는 물환경정보시스템의 최근 10년 수질측정망, 생물측정망 조사자료를 이용해 구축하였다. 피라미의 수질별 서식 적합도는 일반화가법모형(GAM)을 적용하여 수질 인자별 어류 개체 밀도 분포의 상관관계를 분석하여 지수화하였다. 특히, 어류의 서식 특성은 수계별로 상이할 수 있어 가용 데이터의 범위를 고려하여 권역별 수질 인자에 따른 피라미의 서식 특성을 분석하였다. 본 연구로 제시된 권역단위 피라미의 수질 서식처 적합도 지수는 생태를 고려한 하천사업의 계획, 평가의 기초자료를 제시할 수 있을 것이다. 또한, 피라미 외 각 하천의 주요 생태 어종 평가를 위한 수질 서식처 적합도지수 산정의 자동화를 위한 알고리즘 개발에 적용가능할 것으로 예상된다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.23 no.1
• /
• pp.13-25
• /
• 2024

In traffic management, gaps in understanding traffic conditions continue to exist. While the self-belonging problem indicator develops relative to speed, belonging, and self-based relative inclination, it does not apply elimination criteria that may indicate situations that contrast with attribute-specific problems. In this study, we develop integrated indicators that specify communication situations and safety levels for modeling. We review indicators of changes in traffic conditions and raise safety issues, reviewing the indicators so that ITS data can be applied, analyzing the relationships between indicators through factor analysis. We develop combined, integrated indicators that can show changes and stability in traffic situations and that can be applied in traffic information centers to contribute to the development of a traffic environment that can monitor related traffic conditions.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.11-20
• /
• 2024

Among the generative models in Artificial Intelligence (AI), especially Generative Adversarial Network (GAN) has been successful in various applications such as image processing, density estimation, and style transfer. While the GAN models including Conditional GAN (CGAN), CycleGAN, BigGAN, have been extended and improved, researchers face challenges in real-world applications in specific domains such as disaster simulation, healthcare, and urban planning due to data scarcity and unstable learning causing Image distortion. This paper proposes a new progressive learning methodology called Progressive Step Training (PST) based on the Auxiliary Classifier GAN (ACGAN) that discriminates class labels, leveraging the progressive learning approach of the Progressive Growing of GAN (PGGAN). The PST model achieves 70.82% faster stabilization, 51.3% lower standard deviation, stable convergence of loss values in the later high resolution stages, and a 94.6% faster loss reduction compared to conventional methods.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.8
• /
• pp.23-32
• /
• 2024

NVS (Novel View Synthesis) is a field in computer vision that reconstructs new views of a scene from a set of input views. Real-time rendering and high performance are essential for NVS technology to be effectively utilized in various applications. Recently, 3D-GS (3D Gaussian Splatting) has gained popularity due to its faster training and inference times compared to those of NeRF (Neural Radiance Fields)-based methodologies. However, since 3D-GS reconstructs a 3D (Three-Dimensional) scene by splitting and cloning (Density Control) Gaussian points, the number of Gaussian points continuously increases, causing the model to become heavier as training progresses. To address this issue, we propose two methodologies: 1) Gaussian blending, an improved density control methodology that removes unnecessary Gaussian points, and 2) a performance enhancement methodology using a depth estimation model to minimize the loss in representation caused by the blending of Gaussian points. Experiments on the Tanks and Temples Dataset show that the proposed methodologies reduce the number of Gaussian points by up to 4% while maintaining performance.

• Tunnel and Underground Space
• /
• v.34 no.4
• /
• pp.267-282
• /
• 2024

This study aims to develop a wireless communication system for long-term monitoring of high-level radioactive waste disposal facilities. Conventional wired sensors can lead to a deterioration in buffer quality and management difficulties due to the use of cables for power supply and data transmission. This study proposes the adoption of a wireless communication system and compares the received signal strengths within bentonite using modules such as WiFi, ZigBee, and LoRa. Increases in dry density of bentonite and distance between transceivers led to reduced received signal strength. Additionally, using the low-frequency band exhibited less signal attenuation. Based on these findings, a conceptual design for a wireless network-based THM integrated sensor system was proposed. Results of this study can be used as foundational data for long-term monitoring of disposal facility.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.558-569
• /
• 2020

This study examined the frequency of a broiler's stay in each area as stock density using an ICT-based image analysis technique from the perspective of precision livestock farming (PLF) according to the increase in the domestic broiler farms to understand the normal behavior patterns of broilers by age. The broiler was used in the experimental box (3.3×2.7 m) in a poultry house in Gyeonggi province. The stock densities were 9.5 birds/㎡ (n=85) and 19 birds/㎡ (n=170), respectively, and the frequency of stay by feeding, water, and rest area was monitored using a top-view camera. The image data of three-colored-specific broilers identified as the stock density were acquired by age (12, 16, 22, 27, and 29 days) for six hours. In the collected image data, the object tracking technique was used to record the cumulative movement path by connecting approximately 640,000 frames at 30 fps to quantify the frequency of stay in each area. In each stock density, it was significant in the order of the rest area, feeding, and water area (p<0.001). In 9.5 birds/㎡, it was at 57.9, 24.2, and 17.9 %, and 73.2, 16.8, and 10 % in 19 birds/㎡. The frequency of a broiler's stay could be evaluated in each area as the stock density using an ICT-based image analysis technique that minimizes stress. This method is expected to be used to provide basic material for developing an ICT-based management system through real-time monitoring.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1723-1735
• /
• 2022

Landslides are one of the most prevalent natural disasters, threating both humans and property. Also landslides can cause damage at the national level, so effective prediction and prevention are essential. Research to produce a landslide susceptibility map with high accuracy is steadily being conducted, and various models have been applied to landslide susceptibility analysis. Pixel-based machine learning models such as frequency ratio models, logistic regression models, ensembles models, and Artificial Neural Networks have been mainly applied. Recent studies have shown that the kernel-based convolutional neural network (CNN) technique is effective and that the spatial characteristics of input data have a significant effect on the accuracy of landslide susceptibility mapping. For this reason, the purpose of this study is to analyze landslide vulnerability using a pixel-based deep neural network model and a patch-based convolutional neural network model. The research area was set up in Gangwon-do, including Inje, Gangneung, and Pyeongchang, where landslides occurred frequently and damaged. Landslide-related factors include slope, curvature, stream power index (SPI), topographic wetness index (TWI), topographic position index (TPI), timber diameter, timber age, lithology, land use, soil depth, soil parent material, lineament density, fault density, normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used. Landslide-related factors were built into a spatial database through data preprocessing, and landslide susceptibility map was predicted using deep neural network (DNN) and CNN models. The model and landslide susceptibility map were verified through average precision (AP) and root mean square errors (RMSE), and as a result of the verification, the patch-based CNN model showed 3.4% improved performance compared to the pixel-based DNN model. The results of this study can be used to predict landslides and are expected to serve as a scientific basis for establishing land use policies and landslide management policies.

• Journal of Korea Water Resources Association
• /
• v.56 no.2
• /
• pp.103-113
• /
• 2023

The river confluence is a section in which two rivers with different topographical and hyrodynamic characteristics are combined into one, and it is a section in which rapid flow, inflow of sediments, and hydrological topographic changes occur. In the confluence section, the flow of fluid occurs due to the difference in density due to the type of material or temperature difference, which is called a density flow. It is necessary to accurately measure and observe the confluence section including a certain section of the main stream and tributaries in order to understand the mixing behavior of the water body caused by the density difference. A comprehensive analysis of this water mixture can be obtained by obtaining flow field and flow rate information, but there is a limit to understanding the mixing of water bodies with different physical properties and water quality characteristics of rivers flowing with stratigraphic flow. Therefore, this study attempts to grasp the density flow through the water temperature distribution in the confluence section. Among the extensive data of the river, vertical data and water surface data were acquired, and through this, the stratification phenomenon of the confluence was to be confirmed. It was intended to analyze the mixed pattern of the confluence by analyzing the water mixing pattern according to the water temperature difference using the vertical data obtained by measuring the repair volume by installing the ADCP on the side of the boat and measuring the real-time concentration using YSI. This study can supplement the analysis results of the existing water quality measurement in two dimensions. Based on the comparative analysis, it will be used to investigate the current status of stratified sections in the water layer and identify the mixing characteristics of the downstream section of the river.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.7
• /
• pp.848-857
• /
• 2018

Recently, around the world, active development of new renewable energy sources including solar power, waves, and fuel cells, etc. has taken place. Particularly, floating offshore wind farms have been developed for saving costs through large scale production, using high-quality wind power and minimizing noise damage in the ocean area. The development of floating wind farms requires an evaluation of the Maritime Safety Audit Scheme under the Maritime Safety Act in Korea. Floating wind farms shall be assessed by applying the line and area concept for systematic development, management and utilization of specified sea water. The development of appropriate evaluation methods and standards is also required. In this study, proper standards for marine traffic surveys and assessments were established and a systemic treatment was studied for assessing marine spatial area. First, a marine traffic data collector using AIS or radar was designed to conduct marine traffic surveys. In addition, assessment methods were proposed such as historical tracks, traffic density and marine traffic pattern analysis applying the line and area concept. Marine traffic density can be evaluated by spatial and temporal means, with an adjusted grid-cell scale. Marine traffic pattern analysis was proposed for assessing ship movement patterns for transit or work in sea areas. Finally, conceptual design of a Marine Traffic and Safety Assessment Solution (MaTSAS) was competed that can be analyzed automatically to collect and assess the marine traffic data. It could be possible to minimize inaccurate estimation due to human errors such as data omission or misprints through automated and systematic collection, analysis and retrieval of marine traffic data. This study could provides reliable assessment results, reflecting the line and area concept, according to sea area usage.