Search

Severe accident scenarios for nuclear power plants are determined through probabilistic safety analysis (PSA). In this process, it is possible to identify the failure sequence of specific components, but assessing the impact of component failure time on the severity of accident remains a challenge. In this study, a novel approach is presented that utilizes machine learning methodologies such as reinforcement learning (RL) to complement traditional PSA. The proposed process was validated by comparing whether the most severe accident scenarios obtained through critical accident simulations can be reproduced by a RL, since this is a novel use of machine learning techniques. The comparison shows feasibility of exploring critical accident scenarios using RL. To implement the reinforcement learning methodology based on the existing system code, supervised learning model that can predict the remaining time of reactor vessel failure was implemented in this study. Based on this prediction model and data from existing catastrophic accident simulation, RL was implemented. The results obtained from RL were subsequently validated with the results of severe accident code simulation. In summary, new methodology for applying machine learning techniques to the nuclear accident analysis process was presented, and the feasibility and potential of the proposed methodology were discussed. [ABSTRACT FROM AUTHOR]

This paper introduces methods to develop a surrogate model based on deep learning methods and rolling-window forecast for fast and accurate prediction of severe accidents in a nuclear power plant. The surrogate model was trained using time series data, which represents thermal–hydraulic behavior in the nuclear power plant under multi-component failures while various mitigation strategies are also implemented. The model uses a rolling-window forecast to predict selected thermal–hydraulic variables for each time step using the previous time-step variables. To improve the accuracy, the model was further refined to consider the hysteresis effect of the variables using the previous three-time steps. The value of the performance metrics measured by the mean absolute error was reduced by 64 percent in the three-step model compared to the single-step model. The proposed surrogate model has the potential as a practical severe accident simulator for accident management support tools. [ABSTRACT FROM AUTHOR]