The Study on Risk Assessment Method of Severe Accident Management Strategies in Nuclear Power Plants

Shipeng Niu,Yun Yu,Gaopeng Wang,Xinli Yu,Yang Liu,Jiarui Zhang,Yapei Zhang
DOI: https://doi.org/10.1115/icone31-127856
2024-01-01
Abstract:Abstract In order to provide more convenient and efficient severe accident management tools and improve the ability to cope with severe accident conditions, a risk assessment method of severe accident management strategies is proposed based on severe accident analysis, big data and neural network intelligent learning technology. In this method, the critical state parameters of nuclear power plant are selected as the research object, and different severe accident management strategies are adopted at different accident process states for typical severe accident conditions, and a large amount of calculation and analysis are carried out to obtain the changes of critical state parameters after implementing different strategies, and the risk database of severe accident management strategies is developed. Based on the database, the risk assessment model is developed by using neural network intelligent learning, and the model is verified. The results show that the risk assessment model of severe accident management strategies can quickly evaluate the state changes after different severe accident management strategies are adopted under different severe accident conditions, and the error is low, which meets the needs to evaluate the implementation effect of severe accident management strategy. The evaluation results for the critical state parameters will be compared with the setting values, and then the risks of different severe accident management strategies can be evaluated under the severe accident conditions, which can provide technical support for the severe accident management personnel.
What problem does this paper attempt to address?