Influence analysis of alloy elements on irradiation embrittlement of RPV steel based on deep neural network

Abstract

Reactor pressure vessel (RPV) is the most important core equipment in PWR. Its service life determines the service life of nuclear power plant and directly affects the economy and safety of nuclear power plant. Because RPV is serviced at high temperature, high pressure and high energy neutrons for a long time, the properties of RPV steel will significantly degrade, in which irradiation embrittlement is the most important factor for the structural integrity of RPV. In this work, about 700 groups of data such as composition, irradiation conditions and ductile brittle transition temperature of RPV steel are collected, and the data are cleaned and screened for modelling by machine learning. The deep neural network is used for establishing the correlation between key component and irradiation embrittlement of RPV steel. The results show that the lower flux of neutron irradiation will make the radiation embrittlement effect of RPV steel more obvious at the same neutron fluence. Cu, P and Ni are the key factors to influence the △DBTT of RPV steel. The synergistic effect of Cu and Ni on irradiation embrittlement is greater than that of Cu and Mn. These results will help to promote the optimization design of new RPV steel.