Transient identification by clustering based on Integrated Deterministic and Probabilistic Safety Analysis outcomes

In this work, we present a transient identification approach that utilizes clustering for retrieving scenarios information from an Integrated Deterministic and Probabilistic Safety Analysis (IDPSA). The approach requires: (i) creation of a database of scenarios by IDPSA; (ii) scenario post-processing for clustering Prime Implicants (PIs), i.e., minimum combinations of failure events that are capable of leading the system into a fault state, and Near Misses, i.e., combinations of failure events that lead the system to a quasi-fault state; (iii) on-line cluster assignment of an unknown developing scenario. In the step (ii), we adopt a visual interactive method and risk-based clustering to identify PIs and Near Misses, respectively; in the on-line step (iii), to assign a scenario to a cluster we consider the sequence of events in the scenario and evaluate the Hamming similarity to the sequences of the previously clustered scenarios. The feasibility of the analysis is shown with respect to the accidental scenarios of a dynamic Steam Generator (SG) of a NPP.