• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2022년도 봄 학술논문 발표대회
• /
• pp.73-74
• /
• 2022

Concrete structures occupy the largest proportion of modern infrastructure, and concrete structures often have cracking problems. Existing concrete crack diagnosis methods have limitations in crack evaluation because they rely on expert visual inspection. Therefore, in this study, we design a deep learning model that detects, visualizes, and outputs cracks on the surface of RC structures based on image data by using a CNN (Convolution Neural Networks) model that can process two- and three-dimensional data such as video and image data. do. An experimental study was conducted on an algorithm to automatically detect concrete cracks and visualize them using a CNN model. For the three deep learning models used for algorithm learning in this study, the concrete crack prediction accuracy satisfies 90%, and in particular, the 'InceptionV3'-based CNN model showed the highest accuracy. In the case of the crack detection visualization model, it showed high crack detection prediction accuracy of more than 95% on average for data with crack width of 0.2 mm or more.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2019년도 춘계 학술논문 발표대회
• /
• pp.165-166
• /
• 2019

Curtain wall means a non-bearing wall that forms the outer walls of a building to divide the exterior and interior space. The increased use of curtain walls is diverse, including structural safety, watertightness, and wind pressure. As the government's energy conservation policy and the aim of zero-energy houses, the importance of heat reduction is also greatly increased. So, the study of monotony is constantly being conducted. Thus, in this study, insulation performance was analyzed through simulation according to the shape of curtain wall and the shape of insulation inside, and the purpose of this study was to provide basic data on the application of insulation criteria by energy saving design of buildings.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2019년도 춘계 학술논문 발표대회
• /
• pp.163-164
• /
• 2019

Curtain wall is constructed in various forms and designs by forming the exterior walls of a building, and refers to non-bearing walls that divide the exterior and interior spaces of a building. Curtain walls require not only wind pressure, but also waterproof, insulation, insulation, and durability, as well as the effects of recent frequent earthquakes. Studies on the sleeve used to connect the vertical member mullions in the process of high-rise curtain walls are insufficient. In this study, sleeves connecting curtain wall mullions were developed to absorb external displacement, and the purpose of this study was to evaluate economic feasibility through comparison with existing construction methods.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2021년도 학술발표회
• /
• pp.218-218
• /
• 2021

최근 환경 문제과 관련하여 건설 분야에서 발생하는 환경적인 부담을 줄이고자 다양한 재활용 소재들에 대한 연구에 대한 관심이 증가하고 있다. 특히 폐기물 중 생활계에서 발생하는 폐플라스틱은 '쓰레기 대란'이라는 사회적 문제로 대두된 사례가 있다. 생활계에서 발생하는 폐비닐은 다양한 조성으로 이루어져 재활용 대상에서 제외되기 때문에 매립 또는 소각되는 플라스틱의 양을 줄이고 재활용율을 늘릴 수 있는 연구가 필요하다. 본 연구는 생활계에서 발생하는 폐비닐을 융용시켜 무기첨가제 등을 한 복합 Alloy를 생산한다. 또한 재생이 가능한 복합 Alloy를 압출 또는 사출 과정을 통해 테트라포드의 형태를 만들 수 있는 Preform으로 가공하는 공정연구 하였으며 이러한 Preform을 해안 및 항만에서 사용되는 콘크리트 피복재에 적용하였을 때 플라스틱 소재와 콘크리트 소재의 비중의 차이로 인해 발생하는 안정성을 정량적으로 평가하는 수리모형 실험를 수행하였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.796-798
• /
• 2023

IoT 기기 사용량의 증가로 인해 해킹 사례도 함께 증가하며 보안의 중요성이 커지고 있다. 본 논문은 IoT 보안 취약점을 해결하기 위해 정상/악성코드의 데이터셋을 Grayscale로 변환하여 악성코드/정상코드로 분류하는 알고리즘을 개발해 IoT 기기에서 성능을 검증한다. 분류에 이용되는 딥러닝 알고리즘은 CNN(Convolutional Neural Network)으로 99.60%의 평균 정확도를 나타내며 IoT 기기(라즈베리파이)에서도 잘 작동됨을 확인할 수 있다.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2023년도 추계학술대회
• /
• pp.236-237
• /
• 2023

선박무선통신장비는 선박이 항해하는데 필요한 안전정보, 선박교통 모니터링 및 관제, 입·출항 정보를 교환하기 위한 필수 장비이므로 선박항해사는 무선통신 내용을 항상 주의 깊게 청취해야 함. 본 연구에서는 선박의 실제 음성 교신데이터 500시간 데이터를 수집 및 학습하고, Wav2Vec 및 Whisper 모델을 활용하여 한글 및 영어(해사영어) 음성인식 모델을 개발하고 실용화를 수행하였다. 음성인식 모델의 성능은 CER(Character Error Rate) 기준 94.5%로 향후 선박 운항 관련 댜양한 분야에 적용이 가능할 것으로 사료된다.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2023년도 가을학술발표대회논문집
• /
• pp.129-130
• /
• 2023

This study proposes rainwater infiltration retention blocks as a solution to the flooding problems caused by recent climate change and developed a flood prediction simulation program to select the optimal site for installing rainwater infiltration retention blocks that can minimize damage from floods. By applying the existing 2D flood analysis model G2D and adding a reservoir function, the volume of water before and after installation can be determined through simulation results.

• 한국BIM학회 논문집
• /
• 제14권1호
• /
• pp.11-21
• /
• 2024

This study is about a case of web application development using representative commercial BIM platform-linked APIs from Autodesk and Bentley Systems. The development method was aimed at application development using Autodesk's Forge API and Bentley System's iTwin API. The specific development content was limited to the Model Viewer, Issues, and File Transfer API, and the administrator's site. The first research result is that parts with different internal configurations of commercial platforms were independently modularized and developed, and the common front-end was unified. The second research result is that a web front-end technology, was used to visualize 3D models using the viewer API, and the possibility of additional application expansion was discovered.

• 한국산업융합학회 논문집
• /
• 제27권1호
• /
• pp.1-7
• /
• 2024

This study was investigated the effect of heat treatment on the microstructure and hardness of internally hardened ductile cast iron roll. The following conclusions were obtained. Some of the graphite was decreased and a bainite was produced by heat treatment. It decreased due to the decomposition of some of the cementite precipitated in the as-cast by heat treatment, but there was no significant change when it reached a certain depth. Hardness increased due to formation of bainite by heat treatment. On the surface, the hardness decreased due to the decrease in the amount of transformation of cementite into bainite, but there was no change beyond a certain depth.

• 한국전기전자재료학회논문지
• /
• 제37권5호
• /
• pp.494-499
• /
• 2024

The possibility of a dye-sensitized solar cell (DSSC) submodule was evaluated as an independent power source that can drive a smart liquid crystal window (SLW) that selectively blocks sunlight when electricity is applied. In order to save energy and increase the functionality of buildings, SLW operation was supplied directly from DSSC submodule, rather than connecting to the existing power system and external power sources. It was confirmed that the SLW can control light transmittance through self-generation using the DSSC submodule composed of 6 cells at low light of 2,500 lux. These results imply that there is a high possibility of combining smart windows and DSSCs suitable for window-type building-integrated photovoltaic (BIPV) systems. DSSCs, which can self-generate power in low light, are expected to increase their usability in urban BIPV systems through combination with smart window technology.