Uncertainty and Sensitivity Analysis of a Model for Multicomponent Aerosol Dynamics

Abstract

An uncertainty and sensitivity analysis of the MAEROS model for multicomponent aerosol dynamics is presented. Analysis techniques based on Latin hypercube sampling and regression analysis are used to study the behavior of a two-component aerosol in a nuclear power plant containment for a transient accident with loss of alternating current power (i.e., a TMLB’ accident). Conditional on assumed ranges and distributions for selected independent variables (e.g., initial distributions and mass loadings for each component, temperature, pressure, shape factors), estimates are made for distributions of model predictions and for the independent variables that influence these predictions. The analysis indicated that, for the situation under consideration, variables related to agglomeration (e.g., dynamic shape factor, material density, agglomeration shape factor, and turbulence dissipation rate) tended to dominate the observed variability. For comparison, an analysis based on differential techniques is also given. Furthermore, a study of the effects on MAEROS predictions due to the number of particle size classes and the particle size class boundaries is presented. This analysis was performed as part of a project to develop a new system of computer codes (i.e., the MELCOR code system) for use in risk assessments for nuclear power plants.