DATA-DRIVEN ON-LINE PREDICTION OF THE AVAILABLE RECOVERY TIME IN NUCLEAR POWER PLANT FAILURE SCENARIOS
Résumé
This paper presents a data-driven approach for predicting the available Recovery Time (RT) of a system during a failure scenario, i.e., the time remaining until the system can no longer perform its function in an irreversible manner. A library of reference multidimensional trajectory patterns from failure scenarios is created. When a new failure scenario develops, its evolution pattern is compared by fuzzy similarity analysis to the reference multidimensional trajectory patterns. The time remaining before the developing trajectory pattern hits a failure threshold is predicted by combining the times of failure of the reference patterns, weighed by their similarity to the developing pattern. For illustration purposes, a case study of literature is considered regarding the estimation of the available RT in failure scenarios of the Lead Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS).
Origine : Fichiers produits par l'(les) auteur(s)
Loading...