Assessment of Fire Model Uncertainty of Fire Dynamics Simulator in Multiple Compartments of Nuclear Power Plant

It is necessary to consider the uncertainty of a fire model to ensure accuracy in fire risk assessment. Recently, NUREG-1824 reported fire model uncertainty data based on 340 full-scale fire experiment results. However, somewhat inaccurate test results can be obtained when the statistical uncertainty of the model is applied to specific fire risk assessments. In this study, the uncertainty of the fire model, Fire Dynamics Simulator (FDS), was investigated based on PRISME3 fire experiment results. Measurement and prediction results for approximately 200 locations were compared and analyzed, and the quantitative model uncertainty was calculated using the bias factor and relative standard deviation. The results showed that although the computed FDS bias factor of the heat flux based on NUREG-1824 test results was 0.89, 1.18 was obtained in this study, indicating a significant difference. Although the model uncertainties for the same fire experiment results were reviewed, a maximum difference of 12% in model uncertainty was observed, depending on the measurement location.