• Journal of the Korean Society of Systems Engineering
• /
• v.16 no.2
• /
• pp.67-77
• /
• 2020

Traditionally nuclear thermal hydraulic and nuclear safety has relied on numerical simulations to predict the system response of a nuclear power plant either under normal operation or accident condition. However, this approach may sometimes be rather time consuming particularly for design and optimization problems. To expedite the decision-making process data-driven models can be used to deduce the statistical relationships between inputs and outputs rather than solving physics-based models. Compared to the traditional approach, data driven models can provide a fast and cost-effective framework to predict the behavior of highly complex and non-linear systems where otherwise great computational efforts would be required. The objective of this work is to develop an AI algorithm to predict the peak fuel cladding temperature as a metric for the successful implementation of FLEX strategies under extended station black out. To achieve this, the model requires to be conditioned using pre-existing database created using the thermal-hydraulic analysis code, MARS-KS. In the development stage, the model hyper-parameters are tuned and optimized using the talos tool.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2003.11a
• /
• pp.517-523
• /
• 2003

The physical properties of polypropylene (PP) membranes under the radiation field were investigated. To calculate radiation flux affecting to PP, it was used MCNP4A Code. The PP membrane and deoxygenation equipment were standardized to bar structure in order to calculate the phonton flux with MCNP4A Code. The change in the properties of the PP membrane to be used in deoxygenation equipment was rarely occurred during the usage work because the radiation level of reactor coolant water was very low level and The doses of radiation workers are very low. From the results, it was found that the Physical properties of PP membranes which used for nuclear power plant reactor coolant water disposal were not rarely changed under the simulated radiation field.

• The KSFM Journal of Fluid Machinery
• /
• v.16 no.4
• /
• pp.35-43
• /
• 2013

Most of the nuclear power plants(NPPs) adopts pressure difference type flow meters such as venturi and orifice meters for the measurement of feedwater flow rates to calculate reactor thermal power. However, corrosion products in the feedwater deposits on the flow meter by fouling as operating time goes. These effects lead to severe errors in the flow indication and then determination of reactor thermal power. The averaging BDFT, which has developed by Yun et al., has a potentiality to minimize this problem thanks to its inherent measurement principle. Therefore, it is expected that the averaging BDFT can replace the venturi meter for the feedwater pipe of steam generator of NPPs. The present work compares the amplification factor, K, based on CFD calculation against the K obtained from experiments in order to confirm whether a commercial CFD code can be applicable to the evaluation of characteristic for the averaging BDFT. In addition to this, the simulations to take into account of fouling effect are also carried out by rough wall option. The results show that the averaging BDFT is a promising flow meter for the accurate measurement of flow rates in the fouling condition of the NPPs.

• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• 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.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.15 no.1
• /
• pp.45-63
• /
• 2017

The oldest commercial reactor in South Korea, Kori-1 Nuclear Power Plant (NPP), will be shut down in 2017. Proper treatment for decommissioning wastes is one of the key factors to decommission a plant successfully. Particularly important is the recycling of clearance level or very low level radioactively contaminated metallic wastes, which contributes to waste minimization and the reduction of disposal volume. The aim of this study is to introduce a conceptual design of a recycle system and to evaluate the doses incurred through defined work flows. The various architecture diagrams were organized to define operational procedures and tasks. Potential exposure scenarios were selected in accordance with the recycle system, and the doses were evaluated with the RESRAD-RECYCLE computer code. By using this tool, the important scenarios and radionuclides as well as impacts of radionuclide characteristics and partitioning factors are analyzed. Moreover, dose analysis can be used to provide information on the necessary decontamination, radiation protection process, and allowable concentration limits for exposure scenarios.

• Nuclear Engineering and Technology
• /
• v.55 no.9
• /
• pp.3423-3440
• /
• 2023

In this work, a multivariate time-series machine learning meta-model is developed to predict the transient response of a typical nuclear power plant (NPP) undergoing a steam generator tube rupture (SGTR). The model employs Recurrent Neural Networks (RNNs), including the Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid CNN-LSTM model. To address the uncertainty inherent in such predictions, a Bayesian Neural Network (BNN) was implemented. The models were trained using a database generated by the Best Estimate Plus Uncertainty (BEPU) methodology; coupling the thermal hydraulics code, RELAP5/SCDAP/MOD3.4 to the statistical tool, DAKOTA, to predict the variation in system response under various operational and phenomenological uncertainties. The RNN models successfully captures the underlying characteristics of the data with reasonable accuracy, and the BNN-LSTM approach offers an additional layer of insight into the level of uncertainty associated with the predictions. The results demonstrate that LSTM outperforms GRU, while the hybrid CNN-LSTM model is computationally the most efficient. This study aims to gain a better understanding of the capabilities and limitations of machine learning models in the context of nuclear safety. By expanding the application of ML models to more severe accident scenarios, where operators are under extreme stress and prone to errors, ML models can provide valuable support and act as expert systems to assist in decision-making while minimizing the chances of human error.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.17 no.1
• /
• pp.107-120
• /
• 2019

The decommissioning of one nuclear power plant in a multi-unit nuclear power plant (multi-unit NPP) site may pose radiation exposure risk to decommissioning workers. Thus, it is essentially required to evaluate the exposure dose of decommissioning workers of operating multi-unit NPPs nearby. The ENDOS program is a dose evaluation code developed by the Korea Atomic Energy Research Institute (KAERI). As two sub-programs of ENDOS, ENDOS-ATM to anticipate atmospheric transport and ENDOS-G to calculate exposure dose by gaseous radioactive effluents are used in this study. As a result, the annual maximum individual dose for decommissioning workers is estimated to be $2.31{\times}10^{-3}mSv{\cdot}y^{-1}$, which is insignificant compared with the effective dose limit of $1mSv{\cdot}y^{-1}$ for the public. Although it is revealed that the exposure dose of operating multi-unit NPPs does not result in a significant impact on decommissioning workers, closer examination of the effect of additional exposure due to actual demolition work is required. The calculation method of this study is expected to be utilized in the future for planned decommissioning projects in Korea. Because domestic NPPs are located in multi-unit sites, similar situations may occur.

• Journal of Energy Engineering
• /
• v.24 no.3
• /
• pp.96-108
• /
• 2015

A passive containment cooling system has been designed to remove the heat inside a containment during accidents without external power supply. In this work, the PCCS was introduced in the APR1400 plant to replace the containment spray system and, then, the thermal-hydraulic performance of the PCCS was analyzed using the system thermal-hydraulic computer code, MARS. A double-ended cold-leg break accident, which is known to induce the maximum pressure in the containment, is simulated, where the thermal hydraulics of the PCCS, the reactor coolant system, and the containment are simultaneously simulated. The results of the calculations showed that the PCCS can replace the existing spray system and that the containment building and its internal structure also play a very important role for the heat removal during the accident. Some sensitivity calculations were carried out to evaluate the model uncertainty and the effects of design parameters. The limitations of the PCCS are also discussed.