Study of probabilistic safety assessment for a reprocessing plant

According to China's nuclear power development plan, the demand for construction of more large-scale spent fuel reprocessing plant (RP) is increased. Nuclear safety is very important in the whole process of nuclear fuel cycle, and the safety supervision of RPs mainly draws lessons from the safety supervision of nuclear power plants (NPPs). The industry had come to realize that it could not eliminate all risks through its defense-in-depth principle and design basis accident methodologies. All nuclear power countries with advanced reprocessing industry such as the United Kingdom, France and Japan have carried out probabilistic safety assessment (PSA) research for RPs. This paper takes Chinese pilot RP as the research object. First, failure mode and effect analysis (FMEA) and master logic diagram (MLD) methodologies were adopted to identify initiating events (IEs), and then IEs were grouped. After that, fault tree (FT) analysis was performed to various accidents - red oil explosion, fire and criticality. Reliabilities of public systems - power and water and compressed air supply systems were modelled and analysed with GO methodology. And then, consequences of typical accidents were assessed. Finally, application of risk-informed technology in the reprocessing facility was investigated. In brief, PSA for a pilot RP was explored in this paper, and the results could be used as reference for Chinese commercial RP.