The development of Petri net-based continuous Markov chain Monte Carlo methodology applying to dynamic probability risk assessment for multi-state resilience systems with repairable multi-component interdependency under longtermly thereat

For all the nuclear reactor systems, quantitative assessment of the accident management (AM) effects against long-term external hazards became one of the essential issues after the lesson learned from the Fukushima Daiichi Nuclear Power Plant accident. However, the influence from the safety systems' stochastic and dynamic shifting between multiple working states, which is related to the interaction with the adjacent components/systems in general, has not been accounted for yet. Therefore, this research aims to develop a dynamic probability risk assessment tool considering repairable multi-component interdependency for investigating the AM influences on the multi-state safety systems under long-term external hazards. Based on the newly proposed methodology in this research via integrating the Petri net (PN) model with the continuous Markov chain Monte Carlo (CMMC) method, a framework applying PN-CMMC methodology to a severe accident analysis code, SPECTRA, had been originally constructed. Different AM influences on the multi-state decay heat removal systems against long-term volcanic ashfall were also quantitatively confirmed, indicating that halving the repairing time is more influential in suppressing the core damage frequency than doubling the number of adjacent electricity support systems. Therefore, the PN-CMMC-SPECTRA framework can further assess the uncharted dynamic multi-state concerns, leading to a safer AM strategy.