RELAP5/MOD3.2 re-analysis and accuracy quantification of LOFT experiment L2-5

The paper presents the activity performed at University of Pisa in the framework of the participation to the Phase II of the BEMUSE (Best Estimate Methods - Uncertainty and Sensitivity Evaluation) Programme. This activity has been promoted by the Working Group on Accident Management and Analysis (GAMA) and endorsed by the Committee on the Safety of Nuclear Installations (CSNI). The scope of the Programme is to perform Large Break Loss-Of-Coolant Accident (LBLOCA) analyses making reference to experimental data and to a Nuclear Power Plant (NPP) in order to address the issue of 'the capabilities of computational tools' including scaling/uncertainty analysis. The justification for such an activity comes from the consideration that a wide spectrum of uncertainty methods applied to Best Estimate codes exist and are used in research laboratories, but their practicability and/or validity is not sufficiently established to support general use of the codes and acceptance by industry and safety authorities. The consideration of the Best Estimate codes and uncertainty evaluation for Design Basis Accident (DBA), by itself, shows the safety significance of the proposed activity. (author)