RELIABILITY ASSESSMENT OF PASSIVE SAFETY SYSTEMS IN NUCLEAR POWER PLANT

With the continuous development of nuclear power technology, passive safety systems are widely adopted in advanced nuclear power plants. Reliability analysis for passive systems is deferent from active systems, since physical process uncertainty is a significant cause of system failure. This paper proposes a reliability analysis framework for passive safety systems. A practical example is delivered for the case of AP1000 passive residual heat removal system under station blackout accident. Firstly, based on the best-estimate program RELAP5/MOD4.0, a thermalhydraulic(T-H) model with reference to AP1000 is established. Secondly, an artificial neural networks(ANNs) is trained to replace the T-H model in order to improve computational efficiency. Finally, the functional reliability assessment for PRHRS is conducted. The study in this paper can contribute to further research of passive safety systems and system design improvement.