When considering mission times which significantly exceed 24 hours in probabilistic safety analysis, dynamic effects may become important. Even though there exist established dynamic methods to calculate the reliability of multi-train systems for longer mission times, application of these methods on large PSA models of a full scope PSA of a nuclear power plant is often not practicable.
In this paper, a quasi-static method approach for modeling dynamic effects in large PSA models is presented. Rather than remodeling the PSA, potentially significant components are selected for dynamic analysis. For these components, full dynamic analysis is performed in an external model. Based on this analysis, effective probabilities for single failures and common cause failures are calculated which best reflect the results of the dynamic model.
In an example calculation it is shown that this can have a large effect on the PSA results, therefore significantly reducing conservatism in the model.