Uncertainty and Sensitivity Analysis of Phebus FPT-1 Experiment Based on Severe Accident Analysis Code ISAA
Hao Yang,Bin Zhang,Jishen Li,Pengcheng Gao,Zhiran Zhang,Fan Miao
DOI: https://doi.org/10.1115/icone31-136230
2024-01-01
Abstract:Abstract Since the Three Mile Island nuclear accident, international research on severe accidents has advanced through a combination of experimental studies and numerical simulations. The intricate physical processes and chemical phenomena involved in severe reactor accidents have led to a heavy reliance on empirical parameters in existing severe accident analysis programs, generating significant uncertainties. As a result, there is growing interest in applying systematic analysis programs combined with uncertainty quantification methods to evaluate potential severe accident consequences, a topic receiving considerable attention among researchers in severe accident analysis. In 2019, Europe initiated the MUSA project, an international collaboration focusing on severe accident management and uncertainty. The project centers on a series of severe accident uncertainty analysis studies based on the Phebus FPT1 experiment. This serves as a valuable tool for assessing severe accident analysis code effectiveness in simulating severe reactor accident progression and offers an opportunity to analyze and enhance code physical models. Previous numerical analysis studies using the integrated severe accident analysis program ISAA for the FPT experiment revealed the necessity for significant estimations, particularly in core degradation and source term release. This underscores the importance of conducting uncertainty and sensitivity studies on parameters affecting numerical simulation results using advanced uncertainty analysis methods. In this study, we employed a self-developed uncertainty analysis tool to conduct random sampling analysis based on Latin hypercube sampling for various uncertainty parameters, including experimental modeling data and code model parameters. Utilizing the Wilks theory, the study conducted uncertainty quantification studies on the thermal-hydraulic results and hydrogen production in the FPT1 experiment. Data comparison shows the current oxidation model of ISAA overestimates the hydrogen production rate during oxidation, but overall agrees well with experimental measurements. Local sensitivity analysis was also conducted for hydrogen production, further quantifying the impact of uncertainty parameters on numerical simulations. Sensitivity analysis shows the radial radiation heat transfer factor, power, and fuel rod failure criterion significantly impact numerical calculations during loss of coolant accident. A more mechanistic fuel rod failure model is crucial for numerical simulation of severe accident processes. The research not only contributes to refining and improving numerical models within analysis codes but also plays a significant role in reducing uncertainty associated with numerical simulations.