Fire risk analysis based on one-dimensional model in nuclear power plant

Abstract

Fire probabilistic safety assessment (fire PSA) is developed to give the insight of nuclear power plant risk induced by fire accident and the main contributors, and fire accident scenario analysis is one of the important parts in the work to get the key factors such as time to target damage and time to detection. The thermal–hydraulic model simulating fire accident should have enough accuracy and high speed to satisfy the request of fire PSA. Many researches indicate that in a large volume the hot fluid will concentrate in the upper part while the cold fluid will be in the lower part because of density difference under fire condition, that is, the gradients of temperature and of some other parameters in vertical direction are much greater than in horizontal direction. Based on such thermal stratification theory a one-dimensional model is developed, and the buoyant jet is used to simulate the process of heated air flowing up. In this paper a fire in a compartment is analyzed based on one-dimensional model and the temperature distribution is obtained, the results are compared with those of commercial software such as FDT, CFAST and FDS, then the time to target damage is evaluated based on results of different models and the fire non-suppression probabilities are evaluated. The results illustrate that one-dimensional model has better accuracy than FDT and CFAST since such a model can simulate the thermal stratification and natural circulation which exist in the volume simultaneously. Moreover when the fire power is low, the thermal stratification is apparent and air temperature in the hot upper layer is much lower than the critical value of target damage, the one-dimensional model has enough accuracy to be used directly in fire PSA. While the thermal stratification will be weakened when the fire power increases because of the effects of radiation heat transfer and the entrainment by the jet, so more detailed model such as FDS is needed for such situations, however the results of one-dimensional model can give the advice for the proper simulation time of FDS to improve the calculation efficiency when the upper part temperature is close to or higher than the critical value.