• Nuclear Engineering and Technology
• /
• 제55권5호
• /
• pp.1630-1643
• /
• 2023

The correct situation awareness (SA) of operators is important for managing nuclear power plants (NPPs), particularly in accident-related situations. Among the three levels of SA suggested by Ensley, Level 3 SA (i.e., projection of the future status of the situation) is challenging because of the complexity of NPPs as well as the uncertainty of accidents. Hence, several prediction methods using artificial intelligence techniques have been proposed to assist operators in accident prediction. However, these methods only predict short-term plant status (e.g., the status after a few minutes) and do not provide information regarding the uncertainty associated with the prediction. This paper proposes an algorithm that can predict the multivariate and long-term behavior of plant parameters for 2 h with 120 steps and provide the uncertainty of the prediction. The algorithm applies bidirectional long short-term memory and an attention mechanism, which enable the algorithm to predict the precise long-term trends of the parameters with high prediction accuracy. A conditional variational autoencoder was used to provide uncertainty information about the network prediction. The algorithm was trained, optimized, and validated using a compact nuclear simulator for a Westinghouse 900 MWe NPP.

• Nuclear Engineering and Technology
• /
• 제45권4호
• /
• pp.459-468
• /
• 2013

During a station blackout (SBO), the initiating event is a loss of Class IV and Class III power, causing the loss of the pumps, used in systems such as the primary heat transporting system (PHTS), moderator cooling, shield cooling, steam generator feed water, and re-circulating cooling water. The reference case of the SBO case does not credit any of these active heat sinks, but only relies on the passive heat sinks, particularly the initial water inventories of the PHTS, moderator, steam generator secondary side, end shields, and reactor vault. The reference analysis is followed by a series of sensitivity cases assuming certain system availabilities, in order to assess their mitigating effects. This paper also establishes the strategies to mitigate SBO accidents. Current studies and strategies use the computer code of the Integrated Severe Accident Analysis Code (ISAAC) for Wolsong plants. The analysis results demonstrate that appropriate strategies to mitigate SBO accidents are established and, in addition, the symptoms of the SBO processes are understood.

• Nuclear Engineering and Technology
• /
• 제52권1호
• /
• pp.69-77
• /
• 2020

After the Fukushima Daiichi accidents, public concerns on nuclear safety and the corresponding burden of nuclear power plant licensees are increasing. In order to secure public trust and enhance the rationality of current safety regulation, we develop a risk-informed decision making (RIDM) framework for the Korean regulatory body. By analyzing all the regulatory activities for nuclear power plants in Korea, eight action items are selected for RIDM implementation, with appropriate procedures developed for each. For two items in particular - the accident sequence precursor analysis (ASPA) and the significance determination process (SDP) - two customized risk evaluation software has been developed for field inspectors and probabilistic safety assessment experts, respectively. The effectiveness of the proposed RIDM framework is demonstrated by applying the ASPA procedure to 35 unplanned scrams and the SDP to 24 findings from periodic inspections.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2015년도 춘계 종합학술대회 논문집
• /
• pp.439-440
• /
• 2015

There has been a big need of Safety Education for the power plants with a high risk due to the Fukushima Daiichi nuclear disaster and the tragic accident of Sewol Ferry. The object of this research is for studying ways of developing contents for customized Power Plants Safety Education applied with 'Story Viewing' technology in order to improve the present format of Power Plant Safety Education based on hard copied documents so as to prevent human mistakes because of lack of system and ability of initial response which come from safety frigidity shown in the case of Sewol Accident. 'Story-viewing' applied to Power Plant Safety Education is the methodology to enhance information communicability utilizing IT/Visualization technology combined with Story Telling that is an effective propagation way.

• Nuclear Engineering and Technology
• /
• 제53권4호
• /
• pp.1181-1188
• /
• 2021

When abnormal events occur in a nuclear power plant, operators must conduct appropriate abnormal operating procedures. It is burdensome though for operators to choose the appropriate procedure considering the numerous main plant parameters and hundreds of alarms that should be judged in a short time. Recently, various research has applied deep-learning algorithms to support this problem by classifying each abnormal condition with high accuracy. Most of these models are trained with simulator data because of a lack of plant data for abnormal states, and as such, developed models may not have tolerance for plant data in actual situations. In this study, two approaches are investigated for a deep-learning model trained with simulator data to overcome the performance degradation caused by noise in actual plant data. First, a preprocessing method using several filters was employed to smooth the test data noise, and second, a data augmentation method was applied to increase the acceptability of the untrained data. Results of this study confirm that the combination of these two approaches can enable high model performance even in the presence of noisy data as in real plants.

• Nuclear Engineering and Technology
• /
• 제51권3호
• /
• pp.702-708
• /
• 2019

This paper deals with break size estimation of loss of coolant accidents (LOCA) using a nonlinear autoregressive with exogenous inputs (NARX) neural network. Previous studies used static approaches, requiring time-integrated parameters and independent firing algorithms. NARX neural network is able to directly deal with time-dependent signals for dynamic estimation of break sizes in real-time. The case studied is a LOCA in the primary system of Bushehr nuclear power plant (NPP). In this study, number of hidden layers, neurons, feedbacks, inputs, and training duration of transients are selected by performing parametric studies to determine the network architecture with minimum error. The developed NARX neural network is trained by error back propagation algorithm with different break sizes, covering 5% -100% of main coolant pipeline area. This database of LOCA scenarios is developed using RELAP5 thermal-hydraulic code. The results are satisfactory and indicate feasibility of implementing NARX neural network for break size estimation in NPPs. It is able to find a general solution for break size estimation problem in real-time, using a limited number of training data sets. This study has been performed in the framework of a research project, aiming to develop an appropriate accident management support tool for Bushehr NPP.

• Nuclear Engineering and Technology
• /
• 제54권8호
• /
• pp.2960-2973
• /
• 2022

Since Fukushima nuclear power plant (NPP) accident in 2011, the importance of research on various severe accident phenomena has been emphasized. Particularly, detailed analysis of combustion risk is necessary following the containment damage caused by combustion in the Fukushima accident. Many studies have been conducted to evaluate the risk of local hydrogen concentration increases and flame propagation using computational code. In particular, the potential for combustion by local hydrogen concentration in specific areas within the containment has been emphasized. In this study, the process of flame propagation generated inside a reactor core to containment during a loss of coolant accident (LOCA) was analyzed using MELCOR 2.1 code. Later in the LOCA scenario, it was expected that hydrogen combustion occurred inside the reactor core owing to oxygen inflow through the cold leg break area. The main driving force of the oxygen intrusion is the elevated containment pressure due to the molten corium-concrete interaction. The thermal and mechanical loads caused by the flame threaten the integrity of the containment. Additionally, the containment spray system effectiveness in this situation was evaluated because changes in pressure gradient and concentrations of flammable gases greatly affect the overall behavior of ignition and subsequent containment integrity.

• 한국산업정보학회논문지
• /
• 제27권6호
• /
• pp.95-103
• /
• 2022

원자력발전소의 중대사고 발생 시 신속하고 정확하게 사고 상황을 파악해야 하며, 이러한 사고진단 정보를 획득했을 때 적절한 사고관리 및 대응을 수행할 수 있다. 본 연구에서는 국가원자력 재난관리 시스템인 AtomCARE (Computerized technical Advisory system for a Radiological Emergency)로 전송되는 주요 발전소 정보로부터 중대사고 상황을 진단하는데 있어 딥러닝 기술의 접목을 고려하였다. 이를 위하여 주요 시나리오를 선정하고 사고 진행에 따른 상세 시나리오에 대하여 중대사고 해석 코드인 MAAP5 다량 계산을 통한 학습 DB를 구축하였다. 그리고 이 DB의 학습을 통하여 주요 발전소 정보로부터 중대사고 상세 시나리오를 분류할 수 있는, 즉 중대사고 상황을 진단할 수 있는 기술을 개발하였다. 또한 블라인드 테스트와 주성분분석을 통한 검증을 수행하였다. 본 연구에서 개발한 기술은 향후 전체 중대사고 시나리오로 확장 및 적용 가능할 것으로 판단되며 신속하고 정확한 사고진단의 기반기술로 활용 가치가 높을 것으로 기대된다.

• 방사성폐기물학회지
• /
• 제18권2호
• /
• pp.195-206
• /
• 2020

국내 방사능재난대응체계는 일본 후쿠시마 원전 사고와 최근의 국내 대형 재난 교훈 등을 바탕으로 개선돼 오고 있지만, 아직 방사능재난 특성과 후쿠시마 원전 사고 교훈을 완전히 반영하고 있지 못하다. 하나의 방사능재난대응체계에 복수의 국내법이 적용되면서, 실제 상황 시, 대응체계의 실효성을 훼손할 가능성이 있는 법 조항 간 불일치 사항이 존재하기도 한다. 이에 본 논문에서는 방사능재난대응 속성을 분석하고, 방사능재난대응체계 적절성 측면에서 「재난 및 안전관리 기본법」과 「원자력시설 등의 방호 및 방사능 방재 대책법」에서 규정한 방사능재난대응체계, 후쿠시마 원전 사고 시 일본측의 방사능 재난 대응 활동을 분석하고, 이 분석결과를 토대로, 대응체계와 조직 측면에서 국내 방사능재난대응체계 실효성을 제고하기 위해 필요한 개선사항을 도출하여 제시하였다.

• Journal of Radiation Protection and Research
• /
• 제48권3호
• /
• pp.131-143
• /
• 2023

Background: This study examined the detection limit of thyroid screening monitoring conducted at the time of the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident in 2011 using a Monte Carlo simulation. Materials and Methods: We calculated the detection limit of a NaI(Tl) survey meter to measure ¹³¹I accumulation in the thyroid gland of children. Mathematical phantoms of 1- and 5-year-old children were developed in the simulation of the Particle and Heavy Ion Transport code System code. Contamination of the body surface with eight radionuclides found after the FDNPP accident was assumed to have been deposited on the neck and shoulder area. Results and Discussion: The detection limit was calculated as a function of ambient dose rate. In the case of 40 Bq/cm² contamination on the body surface of the neck, the present simulations showed that residual thyroid radioactivity corresponding to thyroid dose of 100 mSv can be detected within 21 days after intake at the ambient dose rate of 0.2 µSv/hr and within 11 days in the case of 2.0 µSv/hr. When a time constant of 10 seconds was used at the dose rate of 0.2 µSv/hr, the estimated survey meter output error was 5%. Evaluation of the effect of individual differences in the location of the thyroid gland confirmed that the measured value would decrease by approximately 6% for a height difference of ±1 cm and increase by approximately 65% for a depth of 1 cm. Conclusion: In the event of a nuclear disaster, simple measurements carried out using a NaI(Tl) scintillation survey meter remain effective for assessing ¹³¹I intake. However, it should be noted that the presence of short-half-life radioactive materials on the body surface affects the detection limit.