Research on Long Term Trend Prediction of Nuclear Power Plants Based on Integrated Framework

Abstract In the current field of nuclear energy, the development of the third generations nuclear power plants (NPPs) are the mainstream, and its safety and economy are the important direction of nuclear energy development. Therefore, the normal, stable and safe operation of the nuclear reactor system is particularly important. In order to improve the safety and stability of the nuclear power system, this paper proposes a life prediction model of a NPP based on a comprehensive framework. In this study, an integrated framework based on BP, LSTM, Informer and other models is built, and the model is verified by the running data of Qinshan simulator. The results show that, compared with the traditional BP and LSTM, the long-term trend prediction model of NPP based on integrated framework has lower MAE and MSE values, indicating that the long-term trend prediction model of NPP based on integrated framework has lower error, which provides strong support for the safe and stable operation of nuclear power system.