Research on Time-Dependent Failure Modeling Method of Integrating Discrete Dynamic Event Tree With Fault Tree
Frontiers in Energy Research
Classical PRA methods such as Fault Tree Analysis (FTA) and Event Tree Analysis (ETA) are characterized as static methods due to predetermined event sequences and success criteria of frontline systems. They are widely accepted for risk analysis of nuclear power plants. Unlike classical PRA, Dynamic PRA (DPRA) couples the stochastic random failures of system with deterministic analysis (by simulation) to determine the risk level of complex systems. It considers the safety significance of the
... ificance of the timing and order of events on accident progression and consequences. However, it is time-consuming to establish a complicated full-scope system simulation model. Meanwhile, thousands of accident scenarios are generated due to randomness of state transition, uncertainty of model and parameters. An overload of modeling, calculating, and post-processing will arise. So, it is a prospective and challenging idea to integrate the classical PRA method with the dynamic PRA method. The objective of this paper is to address an integrated method of risk quantification of accident scenarios. It points out how to treat time-dependent interactions of accident dynamics including random failures, temporal events, configuration changes, and physical process parameters explicitly. Possible dependencies and configuration consistency issues accounting for Discrete Event Tree (DET) branch probabilities are discussed. For DET simulation, some of non-safe-related components to be analyzed could be modeled by FTs for conditional branching probability, instead of a computationally expensive simulation model. A method of integrating FT into DET is introduced which emphasizes on computing the conditional branch probability with FTs online, as well as developing a DET model in case of temporal relations of failure. Finally, a simple case of a Low Pressure Injection System in Large Break Loss-of-Coolant Accident (LBLOCA) scenario is provided as a demonstration.