An Event-Driven Scenario Modeling and Replay Method for Operational Safety Audits in NPPs

In the paper, an event-driven scenario modeling and replay method for operational safety audits in a dynamic human-machine interactive simulation environment is presented. The operational safety audit framework is implemented based on log data mining for scenario data extraction, statechart formalism for human-machine interaction representation, interaction simulator for dynamic event sequence reconstruction and propagation analysis, and state pattern recognition for human error identification in a DiD risk monitor platform. The modeling, simulation and analysis of human factors audit process is demonstrated by a case scenario of loss of both residual heat removal pumps during the refueling outage of nuclear power plants. The validation results show that the event-driven multi-agent simulation approach is effective for log-based discovery of underling human failure events in safety audit management of nuclear power plants.