Modellering av väderberoende deposition av radionuklider för PSA nivå 3

This thesis examined the use of the software AFRY Intelligent Scenario Modelling (AISM) for applications in Level 3 Probabilistic Safety Analysis (PSA). A Gaussian puff atmospheric dispersion model was developed in AISM, using probabilistic meteorological data from the area surrounding the Fukushima Daiichi Nuclear Power Plant (NPP) and the source term from the Fukushima Daiichi accident. The objective of this work was to assess whether the model can predict radionuclide deposition resulting from a radioactive release by comparing simulation results to observed deposition from the Fukushima accident. The model demonstrated the ability to reasonably estimate the order of magnitude of deposition; however, it was unable to accurately predict the angle of deposition. A sensitivity analysis was conducted using the Morris and Sobol’ methods to investigate the influence of meteorological inputs on model results. The Morris method indicated that wind speed and wind direction were the most influential parameters. The application of the Sobol’ method was unsuccessful due to limited computational resources and the high computational cost associated with the method.