• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 2003년도 추계학술발표대회 요약집
• /
• pp.127.1-127.1
• /
• 2003

• 대한교통학회지
• /
• 제30권5호
• /
• pp.3-10
• /
• 2012

이 연구는 회전교차로의 교통사고를 다루고 있다. 연구의 목적은 연립방정식을 이용한 회전교차로의 운전유형별 교통사고모형을 구축하는데 있다. 이를 위해 이 연구는 국내 회전교차로 39개소에서 발생한 148건의 사고자료와 통계 프로그램인 SPSS 17.0을 이용하였다. 또한 사고모형은 2SLS(2단계 최소자승) 추정법을 이용하여 구축하였다. 주요 결과는 다음과 같다. 첫째, 사고건수와 EPDO는 쌍방적 관계를 갖는 것으로 평가되었다. 둘째, 운전유형별로 개발된 6개의 연립방정식은 통계적으로 유의한 것으로 분석되었다. 셋째, 개발된 모형은 공통변수와 특정변수를 사용하여 비교 분석되었다. 마지막으로 독립변수에는 ADT, 상충비, 중차량 비율, 회전차로 폭, 회전차로 수, 접근로 차로폭, 접근로 평균 차로 수, 주차시설 유무 및 정류장 유무가 채택되었다.

• 한국건설안전학회 논문집
• /
• 제2권2호
• /
• pp.76-81
• /
• 2019

무인타워크레인의 운전자격 취득이 쉬워지고 불법개조가 늘어나며 소규모 건설현장에서도 무인타워크레인이 점차 활용되고 있다. 그러나 운전경험이 부족한 타워크레인 운전자의 문제가 커지면서 안전사고도 늘고 작업자도 사고위험에 노출되고 있다. 본 연구에서는 무인타워크레인 사고를 분석하여 사고 원인을 도출하였다. 그 후, 사고 원인의 중요도를 도출하기 위해 AHP 기법을 사용하여 분석하였다. 이 연구의 결과는 무인타워크레인을 사용하는 사업자와 건설 노동자들이 작업 규칙을 따르지 않는다는 것이다.

• 농업과학연구
• /
• 제44권4호
• /
• pp.620-635
• /
• 2017

This study was intended to identify the main factors responsible for the decline in purchase of imported agricultural and fish products after Japan's nuclear power plant accident in 2011 and to compare the effects on imported agricultural produce and imported fish products. Logit model and multiple regression model analyses were performed using consumers' survey data. Psychological and qualitative factors reflecting consumers' food safety awareness and purchasing preferences, which were extracted by Factor analysis, were included as the models' explanatory variables, along with socio-demographic and economic factors. The Logit estimation showed aged, married, and low-income households had significantly higher probability of reducing their purchases of imported agricultural and fish products. However, the multiple regression results pointed out that the actual rate of decrease of imported agricultural and fish products purchases were more significantly affected by non-socio demographic factors such as past experience of purchasing imported agricultural and fish products, future intention to purchasing Japanese agricultural and fish products, and the ratio of imported to domestic agricultural and fish products before the nuclear accident, as well as consumers' feeling of food insecurity and their purchasing preferences. Moreover, the results showed that Korean consumers have reacted more sensitively to the decline in imported fish products than imported agricultural produce after the nuclear accident based on the marginal effects of various socio-demographic and economic factors.

• Nuclear Engineering and Technology
• /
• 제45권6호
• /
• pp.767-790
• /
• 2013

MAAP4 is a computer code that can simulate the response of a light water reactor power plant during severe accident sequences, including actions taken as part of accident management. The code quantitatively predicts the evolution of a severe accident starting from full power conditions given a set of system faults and initiating events through events such as core melt, reactor vessel failure, and containment failure. Furthermore, models are included in the code to represent the actions that could mitigate the accident by in-vessel cooling, external cooling of the reactor pressure vessel, or cooling the debris in containment. A key element tied to using a code like MAAP4 is an uncertainty analysis. The purpose of this paper is to present a MAAP4 based analysis to examine the sensitivity of a key parameter, in this case hydrogen production, to a set of model parameters that are related to a Level 2 PRA analysis. The Level 2 analysis examines those sequences that result in core melting and subsequent reactor pressure vessel failure and its impact on the containment. This paper identifies individual contributors and MAAP4 model parameters that statistically influence hydrogen production. Hydrogen generation was chosen because of its direct relationship to oxidation. With greater oxidation, more heat is added to the core region and relocation (core slump) should occur faster. This, in theory, would lead to shorter failure times and subsequent "hotter" debris pool on the containment floor.

• 대한안전경영과학회지
• /
• 제24권4호
• /
• pp.41-47
• /
• 2022

A reliable prediction model of national occupational accident fatality rate can be used to evaluate level of safety and health protection for workers in a country. Moreover, the socio-economic aspects of occupational accidents can be identified through interpretation of a well-organized prediction model. In this paper, we propose a machine learning based relative comparison methods to predict and interpret a national occupational accident fatality rate based on socio-economic indicators. First, we collected 29 years of the relevant data from 11 developed countries. Second, we applied 4 types of machine learning regression models and evaluate their performance. Third, we interpret the contribution of each input variable using Shapley Additive Explanations(SHAP). As a result, Gradient Boosting Regressor showed the best predictive performance. We found that different patterns exist across countries in accordance with different socio-economic variables and occupational accident fatality rate.

• 한국정보기술학회 영문논문지
• /
• 제10권2호
• /
• pp.167-175
• /
• 2020

In this paper, we propose a traffic accident detection on vehicle-mounted camera. In the proposed method, the minimum bounding box coordinates the central coordinates on the bird's eye view and motion vectors of each vehicle object, and ego-motions of the vehicle equipped with dash-cam are extracted from the dash-cam video. By using extracted 4 kinds features as the input of Bi-LSTM (bidirectional LSTM), the accident probability (score) is predicted. To investigate the effect of each input feature on the probability of an accident, we analyze the performance of the detection the case of using a single feature input and the case of using a combination of features as input, respectively. And in these two cases, different detection models are defined and used. Bi-LSTM is used as a cascade, especially when a combination of the features is used as input. The proposed method shows 76.1% precision and 75.6% recall, which is superior to our previous work.

• Nuclear Engineering and Technology
• /
• 제44권5호
• /
• pp.459-470
• /
• 2012

From the PSA point of view, the Fukushima accident of Japan in 2011 reveals some issues to be re-considered and/or improved in the PSA such as the limited scope of the PSA, site risk, etc. KAERI (Korea Atomic Energy Research Institute) has performed researches on the development of an integrated risk assessment framework related to some issues arisen after the Fukushima accident. This framework can cover the internal PSA model and external PSA models (fire, flooding, and seismic PSA models) in the full power and the low power-shutdown modes. This framework also integrates level 1, 2 and 3 PSA to quantify the risk of nuclear facilities more efficiently and consistently. We expect that this framework will be helpful to resolve the issue regarding the limited scope of PSA and to reduce some inconsistencies that might exist between (1) the internal and external PSA, and (2) full power mode PSA and low power-shutdown PSA models. In addition, KAERI is starting researches related to the extreme external events, the risk assessment of spent fuel pool, and the site risk. These emerging issues will be incorporated into the integrated risk assessment framework. In this paper the integrated risk assessment framework and the research activities on the emerging issues are outlined.

• 시스템엔지니어링학술지
• /
• 제18권2호
• /
• pp.94-107
• /
• 2022

Accidents prevention and mitigation is the highest priority of nuclear power plant (NPP) operation, particularly in the aftermath of the Fukushima Daiichi accident, which has reignited public anxieties and skepticism regarding nuclear energy usage. To deal with accident scenarios more effectively, operators must have ample and precise information about key safety parameters as well as their future trajectories. This work investigates the potential of machine learning in forecasting NPP response in real-time to provide an additional validation method and help reduce human error, especially in accident situations where operators are under a lot of stress. First, a base-case SGTR simulation is carried out by the best-estimate code RELAP5/MOD3.4 to confirm the validity of the model against results reported in the APR1400 Design Control Document (DCD). Then, uncertainty quantification is performed by coupling RELAP5/MOD3.4 and the statistical tool DAKOTA to generate a large enough dataset for the construction and training of neural-based machine learning (ML) models, namely LSTM, GRU, and hybrid CNN-LSTM. Finally, the accuracy and reliability of these models in forecasting system response are tested by their performance on fresh data. To facilitate and oversee the process of developing the ML models, a Systems Engineering (SE) methodology is used to ensure that the work is consistently in line with the originating mission statement and that the findings obtained at each subsequent phase are valid.

• 한국화재소방학회논문지
• /
• 제33권6호
• /
• pp.72-79
• /
• 2019

본 연구를 통해 화학사고 사상사고 예측모형을 개발하였다. 모형은 로지스틱회귀분석 모델을 활용하여 사상사고에 영향을 주는 변수를 도출하여 적용하였고, 통계적 검증방법과 오즈비를 활용하여 모형의 신뢰성 및 정확성을 검증하였다. 모형에 활용한 사고 자료는 과거 발생했던 화학사고 통계를 분석하여 활용하였으며, 사고의 유형, 원인, 발생 장소, 사상자 현황 및 사상자를 발생시킨 화학사고 등의 자료 분석을 통해 통계적으로 유의하게 나타난 독립변수(p < 0.05)를 적용하였다. 본 연구에서 개발한 모형은 사업장에서 화학사고로 인해 발생하는 사상사고의 예방 및 안전시스템 구축을 위한 연구로서 의의가 있다고 할 수 있다. 모형에 의한 분석결과 사상사고 발생에 가장 크게 영향을 미치는 변수는 폭발에 의한 화학사고인 것으로 조사되었다. 따라서 사업장에서 발생하는 폭발 유형의 화학사고를 예방하기 위한 대책마련이 시급하다고 판단된다.