The maturity of severe accident (SA) codes, progress in computational methods, and computer infrastructures were considered a sound platform for conducting, for the first time in SA modeling, a systematic and broad application of uncertainty and sensitivity analysis (UaSA) in this domain. The overall objective of the HORIZON2020 project on “Management and Uncertainties of Severe Accidents (MUSA)” was to quantify the uncertainties of SA integral codes when modeling reactor and spent fuel pool (SFP) accident scenarios of Generation-II and Generation-III reactor designs for the prediction of the radiological source term. This paper presents the major achievements accomplished by MUSA. To name a few, (1) a database on input parameter uncertainties has been assembled, (2) SA codes and statistical tools, either inhouse built or existing ones, have been properly coupled by scripting or interfaces, (3) major specific challenges have been identified and different solutions have been worked out, and (4) applications to several reactor types and SFP scenarios have shown interesting insights into this simulation approach, particularly when dealing with source term variables as figures of merit. No less important, major challenges were found and are discussed here as issues that need to be addressed before achieving an effective implementation of UaSA in the SA domain.
Read full abstract