Risk Prediction Of Nuclear Power Plant Based On Equipment Health Status

The nuclear power plant risk monitor system(RM)based on probabilistic safety analysis technology can determine the instantaneous risk of the nuclear power plant and execute configuration risk management according to the actual availability or unavailability of the system and equipment.During the long-term operation of nuclear power plants,there are situations where equipment is available but its performance is degraded.However,since the current risk monitor is based on a general database and assumes that the failure rate of equipment remains constant,it is impossible to monitor the impact of equipment performance degradation on the risk of nuclear power plants.Prognostics and health management technology(PHM)can predict the future state and remaining life of equipment,and the combination of PHM and RM can effectively make up for the shortcomings of traditional risk monitors.Based on the real-time online risk monitoring technology of nuclear power plant,this paper carries out the research of the risk prediction technology scheme combining RM and PHM,and constructs the risk prediction and health management framework combining them.Secondly,this paper uses PHM to obtain the equipment’s health status and remaining useful life information,adopts Bayesian prior distribution and posterior distribution algorithm to update the occurrence frequency of the initial event,and analyzes the steam generator heat transfer tube rupture accident as a case.In terms of updating the basic event failure probability,this paper integrates the equipment failure rate function to obtain the changing information of the failure probability over time.Meanwhile,the impact of preventive maintenance on the equipment failure rate is also considered,and this paper analyzes the failure accident of the heat transfer system of a heavy water reactor as a case study.The basic event failure probability and the time changing data of the HWR heat transfer system failure probability after preventive maintenance are calculated.On this basis,taking a steam generator heat transfer tube rupture accident in Fuqing PWR as an example,the change of core failure frequency with time is predicted under the premise that both the initiation event and the residual heat removal system are affected by equipment deterioration.Finally,in order to make up for the shortcomings of traditional risk monitor software,this paper designs a risk forecasting module,and analyzes the main functions required by the risk forecasting module from two aspects: the risk forecasting process and the risk forecasting index system.And according to the above requirements,the risk prediction module interface is briefly introduced,including two parts: the frequency interface of initiating events and the failure probability interface of basic events.