• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.12
• /
• pp.1637-1639
• /
• 2014

In the past, the analog protection relays have been widely used for the safety-related systems in the nuclear power plants due to their stability and reliability. Meanwhile, as the high performance digital system has been developed, the digital systems have been adopted in the non-safety systems. However, since the digital systems currently used in the non-safety systems were not developed according to Q-class standard, Commercial Grade Item Dedication (CGID) procedure should be performed in order to apply them to the safety-related system. The purpose of this paper is to describe the CGID procedure including the analysis of the hardware architecture as well as the software embedded in protective relay to apply to the emergency diesel generator in the nuclear power plant. The entire CGID procedure was performed strictly according to the international standard and regulations.

• Journal of the Korean Society of Systems Engineering
• /
• v.16 no.2
• /
• pp.1-15
• /
• 2020

Nuclear power plants in South Korea are operated to cover the baseload demand. Hence they are operated at 100% rated power and do not deploy power tracking control except for startup, shutdown, or during transients. However, as the contribution of renewable energy in the energy mix increases, load follow operation may be needed to cover the imbalance between consumption and production due to the intermittent nature of electricity produced from the conversion of wind or solar energy. Load follow operation may be quite challenging since the operators need to control the axial power distribution and core reactivity while simultaneously conducting the power maneuvering. In this paper, a systems engineering approach for multi-physics load follow simulation of APR1400 is performed. RELAP5/SCDAPSIM/MOD3.4/3DKIN multi-physics package is selected to simulate the Korean Advanced Power Reactor, APR1400, under load follow operation to reflect the impact of feedback signals on the system safety parameters. Furthermore, the systems engineering approach is adopted to identify the requirements, functions, and physical architecture to provide a set of verification and validation activities that guide this project development by linking each requirement to a validation or verification test with predefined success criteria.

• Nuclear Engineering and Technology
• /
• v.53 no.12
• /
• pp.4014-4021
• /
• 2021

If safety injection systems (SISs) do not work in the event of a loss-of-coolant accident (LOCA), the accident can progress to a severe accident in which the reactor core is exposed and the reactor vessel fails. Therefore, it is considered that a technology that provides recoverable maximum time for SIS actuation is necessary to prevent this progression. In this study, the corresponding time was defined as the golden time. To achieve the objective of accurately predicting the golden time, the prediction was performed using the deep fuzzy neural network (DFNN) with rule-dropout. The DFNN with rule-dropout has an architecture in which many of the fuzzy neural networks (FNNs) are connected and is a method in which the fuzzy rule numbers, which are directly related to the number of nodes in the FNN that affect inference performance, are properly adjusted by a genetic algorithm. The golden time prediction performance of the DFNN model with rule-dropout was better than that of the support vector regression model. By using the prediction result through the proposed DFNN with rule-dropout, it is expected to prevent the aggravation of the accidents by providing the maximum remaining time for SIS recovery, which failed in the LOCA situation.

• Nuclear Engineering and Technology
• /
• v.56 no.9
• /
• pp.3914-3924
• /
• 2024

The release source term of radioactivity becomes a critical foundation for emergency response and accident consequence assessment after a nuclear accident Rapidly and accurately inverting the source term remains an urgent scientific challenge. Today source term inversion based on meteorological data and gamma dose rate measurements is a common method. But gamma dose rate actually includes all nuclides information, and the composition of radioactive nuclides is generally uncertain. This paper introduces a novel nuclear accident source term inversion model, which is Improve Snow Ablation Optimizer-Sensitivity Analysis Pruning Extreme Learning Machine (ISAO-SAELM) model. The model inverts the release rates of 11 radioactive nuclides (I-131, Xe-133, Cs-137, Kr-88, Sr-91, Te-132, Mo-99, Ba-140, La-140, Ce-144, Sb-129). It does not require the use of the physical field of the reactor to obtain prior information and establish a dispersion model. And the robustness is validated through noise analysis test. The mean absolute errors of the release rates of 11 nuclides are 15.52 %, 15.28 %, 15.70 %, 14.99 %, 14.85 %, 15.61 %, 15.96 %, 15.42 %, 15.84 %, 15.13 %, 17.72 %, which show the significant superiority of ISAO-SAELM. ISAO-SAELM model not only achieves notable advancements in accuracy but also receives validation in terms of practicality and feasibility.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.799-806
• /
• 2024

Construction planners for hard rock tunnel projects often encounter practical challenges caused by inherent uncertainties in ground conditions and resource constraints. Therefore, planners cannot rapidly generate optimal excavation schedules for the shortest project durations with a given equipment fleet by considering the uncertainties in ground conditions. Although some schedule optimization methods exist, they are not tailored for resource-constrained hard rock tunnel projects. To overcome these limitations, the authors specified a formal Q-learning-based schedule optimization methodology for resource-constrained hard rock tunnel projects. States are defined according to the locations of tunnel faces under excavation. Actions consist of multiple and comprehensive heuristic-based rules, which are efficient methods for resource allocation. Rewards are the time intervals required between current states and next states. After that, the methodology is validated using a case study. The generated Q tables indicate (1) best actions under different states and (2) the shortest remaining durations when the project starts from specific (state, action) pairs. The results demonstrate that the optimal schedules can be obtained by applying the proposed methodology. Furthermore, it is beneficial for planners to rapidly assign optimal rules for each state under one ground condition scenario. The results further show the potential to consider the uncertainties in ground conditions using the information of possible ground condition scenarios provided.

• Progress in Medical Physics
• /
• v.30 no.2
• /
• pp.49-58
• /
• 2019

Purpose: Automated analytical systems have begun to emerge as a database system that enables the scanning of medical images to be performed on computers and the construction of big data. Deep-learning artificial intelligence (AI) architectures have been developed and applied to medical images, making high-precision diagnosis possible. Materials and Methods: For diagnosis, the medical images need to be labeled and standardized. After pre-processing the data and entering them into the deep-learning architecture, the final diagnosis results can be obtained quickly and accurately. To solve the problem of overfitting because of an insufficient amount of labeled data, data augmentation is performed through rotation, using left and right flips to artificially increase the amount of data. Because various deep-learning architectures have been developed and publicized over the past few years, the results of the diagnosis can be obtained by entering a medical image. Results: Classification and regression are performed by a supervised machine-learning method and clustering and generation are performed by an unsupervised machine-learning method. When the convolutional neural network (CNN) method is applied to the deep-learning layer, feature extraction can be used to classify diseases very efficiently and thus to diagnose various diseases. Conclusions: AI, using a deep-learning architecture, has expertise in medical image analysis of the nerves, retina, lungs, digital pathology, breast, heart, abdomen, and musculo-skeletal system.

• Journal of the Korean Society of Safety
• /
• v.29 no.2
• /
• pp.62-67
• /
• 2014

The purpose of this study is to analyze the degree of risk of the working unit of nuclear power plants construction. In order to do this, and the risk index by type and source of risk judgment derived in the previous study were utilized. Further, to derive a risk index of unit work in nuclear power plant construction, a survey targeting safety professionals was conducted. The analytic hierarchy process (AHP) was used for analysis of the survey. The following results were obtained. Firstly, the results of AHP showed that main building work is the most dangerous work, and base excavation work is the second dangerous work among 21 unit works. Secondly, so, it is required to invest more and to take a increasing interest in unit works of civil and architecture as compared to other unit works. Further, the results could be used to reduce the degree of risk in construction of the nuclear power plant.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.16 no.3
• /
• pp.55-62
• /
• 2013

Generally, the radioactivity from NPP(Nuclear Power Plants) operation can be released below 3% of DRLs(Derived Release Limits) to environment. It was tried to understand which plant was efficient for absorbing radioactivity in this study. Pinus thunbergii Parl. and Viburnum awabuki K. KOCH were analyzed for radioisotope absorption. The samples were collected at three different locations depending on the distance from NPP at the vicinity 10km away, and 30km away. Gamma radionuclide was not detected from the samples, which means that the direct transition into the plant was not significant. Meanwhile, the very low level of radioactive tritium was detected in the samples. One remark was that every plant has different ability for tritium absorption. These results are expected to be applied to propagation and transplanting in radioactively contaminated area or reducing radioactivity in the soil and water near the plants.

• Journal of the Korean Society of Safety
• /
• v.28 no.3
• /
• pp.95-99
• /
• 2013

The goal of this study was to investigate some significant factors to influence level of safety at plant construction field and analyze degree of risk by work classification. Currently, there are lots of construction fields for the nuclear power plant for electricity generation, and our government also planned constructing more nuclear power plant in near future. However, much of the safety literature neglected the degree of risk factors on the plant construction field. Safety managers participated in the brainstorming session for drawing decision criteria of the degree of risk (i.e., significant factors). Then, they were asked to answer a structured questionnaire which was developed for drawing most important factors. Finally, the analytic hierarchy process (AHP) was used to analyze level of risk by work classification. The following results were obtained. First, total twelve factors judging degree of risk were found in the brainstorming session. Second, the questionnaire showed four significant factors, including number of workers, working environments, skill of craft and accident experience. Third, the results of AHP showed Architecture work is the most dangerous work among 6 work types. The results could be used to reduce degree of risk in construction field of the nuclear power plant.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• 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.