• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.667-669
• /
• 2022

최근 고도화 되고 있는 인공지능 기술은 복잡한 데이터 속에서 내재된 인사이트를 발견하여 상태 변화를 진단하고 예측하는데 활용되고 있다. 이러한 첨단 기술을 활용하여, 원자력 발전소에서 공정상태가 비정상 또는 비상 상태로 악화되기 전에 운전원이 인지할 수 있다면 공정상태를 정상으로 회복하는데 도움을 줄 수 있다. 이 논문에서는 공정상태 판단을 위한 딥러닝 모델을 활용하는 지능형 조기 경고 개념을 제안한다. 공정상태의 변화 추세를 예측하는 목적으로 사용하는 지능형 조기 경고는 기존 경보 보다 단순화된 상태 천이 메커니즘을 사용하여 운전원의 부담이 증가하지 않도록 한다. 또한, 사고 시나리오 데이터로 딥러닝 모델로 학습하고 지능형 조기 경고 화면을 구현하여 운전원을 지원하기 위한 구현방향을 제시한다.

• Fire Science and Engineering
• /
• v.20 no.2 s.62
• /
• pp.21-30
• /
• 2006

Since Daegu subway arson fire disaster in 2003, there have been several copycat crimes such as at Seoul Metro line number seven and Hongkong Chuiwan line cases. Oil and gas were used for fire propagation in most cases as in Daegu case and such fire could be expanded to a whole subway cabin within several minutes. The fire may eventually cause the whole subway system stop. Fire damage can be minimized when fire occurrence and diffusion are blocked by stages or isolated rapidly. This study suggests an effective early response system that separates passengers from fire and a real-time fire extinguishment program by stages. Based on the subway arson case studies, the early response scenario has been structured by three stages, i.e., confirmation of fire and damage, early fire extinguishment, and information dissemination and passengers evacuation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.3
• /
• pp.488-496
• /
• 2020

This study was based on nitrate chemical accidents at home and abroad. Toxic gases due to adverse reactions are generated in the workplace, laboratory, hospital, container damage, and chemical misinjection. Through a case review of possible situations and safety, this study analyzed various cases of accidents, accident status, accident type, cause of the accident, location of the accidents, etc. from 2014 to 2018. The plans for improvement in education and nitrate accidents were reviewed. As a result, 36 nitrate chemical accidents were investigated, including 16 careless worker accidents, eight transportation accidents, and 12 facilities shortages. Nitrate chemical accidents are occurring continuously. Based on this, the range of toxic effects using CARIS was calculated at the worst-case scenario, and the effective response range was measured through the damage impact range. For this purpose, the impact range was predicted based on the strengthening of safety education, emergency action plan and correlation, and the quantified data was identified. In addition, the reliability of the scope of impact was reviewed based on the correlation formula that could facilitate the evacuation of residents, and it was applied to actual accident scenarios of the workplace to present the effects of the accident response and preventive measures.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.4
• /
• pp.485-493
• /
• 2009

The u-City offers diverse u-services, which are integrated with information and GIS technologies, by utilizing an integrated management platform for controlling the city efficiently, increasing convenience of city lift, and improving the quality of lives. An urban control system is a core element of the u-City to connect organically and expand all kinds of information through various u-interfaces. This study presents the implementation of a GIS service module based on the urban control system for the efficient development of u-services, which contributed to the successful execution of GIS service scenarios that include processes of unexpected events, air pollution monitoring, and emergences. The results of the study can be used as basic data for the design and operation of practical u-City services.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.39-39
• /
• 2017

한반도 기후변화 경향은 이미 기상 생태 환경 수자원 등 광범위한 부분에서 감지되고 있다(기상청, 2011a, 2011b). 현재까지의 연구에 따르면 한반도 기후변화에 따른 영향으로 강우패턴은 첨두강우가 7월에서 점차 8월로 이동 변화하는 것으로 전망되고, 양적으로 연강수량은 점차 감소가 전망되면서도 극한 값은 발생빈도와 크기가 점증할 것으로 전망되고 있다. 그래서 그간 기존댐에 대한 재평가(1998, 2010, 2012)와 발생 가능한 최대강우량 설계기준으로 기존 댐의 여수로 배제 능력을 증대시키는 비상여수로 설치 등 기존 댐 시설위주의 효율적인 기후변화 대응 또는 적응 방안을 시행해 왔다. 그러나, 기후변화로 인한 기상 상황은 전에 발생한 적이 없었던 새로운 기상이변과 재난을 가져오고 있다. 이에 기상 변화에 하나의 시설로서 대응 하던 방식에서 한 번의 기상이변이 유역 전반에 걸쳐 재난을 발생하는 최근의 상황에 맞추어 수자원 시설을 계획하는 방식에 대한 변화의 필요성 있다고 생각하였다. 이에 장래 전망되는 기후변화를 감안하여 이수와 치수 시설의 가뭄과 홍수에 대한 대처 능력을 유역 차원에서 평가하는 방법을 찾아보고자 한다. Robust 하다는 것은 어떤 상황에서도 작동이 되는 것을 말하는 강건한 계획으로, 이와 같은 시설 계획을 위해서는 먼저 현재의 시설물에 대한 회복력을 판단하는 평가가 있어야 할 것이다. 따라서, 용수공급이든 홍수 재난이든 회복력(복원력)에 대한 평가를 하고, 대안에 대한 로버스트 의사 결정 방법(RDM: Robust Decision Making)을 적용하여 우수한 대안을 찾으면 강건한 시설계획 수립이라는 절차가 될 수 있다고 판단하였다. 본 연구는 회복력(복원력)을 갖는 로버스트 의사결정방법에 대한 과거 연구 조사를 기초로 하여 연구 수행 절차를 마련한 후에 장래 한반도 기후변화 시나리오를 시범 유역에 적용하여 수자원 시설의 복원 또는 회복력을 분석하고, robust 의사결정방법을 적용함으로써, 향후 로버스트 수자원 시설 계획이 어떻게 이루어져야 하는지와 함께 이수와 치수 시설의 종합적인 계획 등에 대한 개념적인 절차와 방법의 제시를 도모하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.370-370
• /
• 2022

최근 우리나라는 수자원부존량 대비 수자원이용률이 16%에 불과하며 평상시 하천의 많은 가용수량이 이용되지 못하고 있는 실정이다. 현재 하천수 사용은 갈수와 가뭄시를 기준으로 허가 및 관리하고 있으며, 평상시 가용수량을 모니터링하여 하천유지유량, 하천수 사용, 환경대응 등 다양한 수요에 맞게 수량을 합리적으로 배분할 수 있는 수자원관리 체계 구축이 필요한 실정이다. 이에 본 연구에서는 대상지역인 금강유역의 하천수 사용 및 공급 상황을 파악하고 혼합(유역·하도) 물수지를 이용하여 실시간 가용수량 평가체계를 구축하였다. 실시간 가용수량 평가체계를 구축하기 위해서 유역의 수문상황과 기상전망(기온, 강우) 자료를 이용하여 장래기간의 강우조건(무강우, Ensemble Streamflow Prediction; ESP), 하천수 사용 및 공급계획(댐, 저수지 방류량 등) 시나리오 조건에 따른 예측기반 가용수량을 평가하였다. 대상 하천의 지류 유입량을 산정하기 위해서는 유역 물수지를 수행하고 하도 물수지를 통해 주요 지점별 하천관리유량과 가용수량을 산정한다. 또한, 유역(수계)간 물이동 및 하천수 사용 시설별 용·배수 체계를 고려하여 물수지 분석을 수행하였다. 이를 통해 산정된 각 취수시설별 유입량, 사용량, 회귀수량, 하천유지유량, 하천관리유량, 유량, 가용수량 등을 사용자가 쉽게 확인할 수 있도록 GIS와 테이블 기반으로 표출하였다. 본 평가체계를 활용하여 홍수통제소에서 수원(저수, 유수)별 가용수량을 다양한 수요에 맞게 적절하게 배분하고 조정할 수 있고 추후 가뭄 등 비상상황 발생 시 용수공급조정 및 연계운영계획에 반영될 수 있을 것으로 판단된다.

• Aerospace Engineering and Technology
• /
• v.11 no.2
• /
• pp.129-139
• /
• 2012

This paper will describe the structure and characteristics of operator training system which was developed to maintain the quality of operational ability for COMS (Communication, Ocean and Meteorological Satellite) operators during a long-term nominal operations of missions and a contingency operations against an occurrence of anomaly. In particular it will present benefits and expected effects of the training system with a focus on three parts which are functions especially for trainer-friendly failure injection, an automatic sequencer of training scenario based on the predefined plan and additional functions of the existing COMS simulator. Furthermore, it will present a practical example of training on the training system to understand the overall mechanism of training process.

• Journal of Navigation and Port Research
• /
• v.40 no.4
• /
• pp.189-196
• /
• 2016

We propose a virtual sailor training platform which supports emergency drills for ship's fire in virtual environment. Proposed platform not only enhances training efficiency by providing immersiveness, but also enables a consolidated virtual training due to the network-based multiplayer capabilities. Based on the offline fire simulation results using FDS and CFAST the platform visualizes a realistic fire spread in real-time. The training platform on the basis of the fire training material of the maritime safety education institute induces equipment proficiency and environment adaptation throughout immersive virtual environment in addition to procedure proficiency as well. In the implementation we showed that the equipment and environment controls and telepresence improve the training proficiency and enable collaborative virtual training that participates multiple trainees and induces cooperation for a common goal. Implementation of the platform demonstrated the skill mastery capability of the drill such as efficient fire apparatus controls and passenger controls.

• Journal of Internet Computing and Services
• /
• v.17 no.6
• /
• pp.1-8
• /
• 2016

Current indoor intrusion detection and location tracking methods have the weakness in seamless operations in tracking the objective because the object must possess a communicating device and the limitation of the single cell size (approximate $100cm{\times}100cm$) exits. Also, the utilization of CCTV technologies show the shortcomings in tracking when the object disappear the area where the CCTV is not installed or illumination is not enough for capturing the scene (e.g. where the context-awarded system is not installed or low illumination presents). Therefore, in this paper we present an improved in-door tracking system based on sensor networks. Such system is built on a simulated scenario and enables us to detect and extend the area of surveillance as well as actively responding the emergency situation. Through simulated studies, we have demonstrated that the proposed system is capable of supplementing the shortcomings of signal cutting, and of estimating the location of the moving object. We expect the study will improve the better analysis of the intruder behavior, the more effective prevention and flexible response to various emergency situations.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.