This paper is aimed at describing results of the first part of the BEMUSE (Best Estimate Methods – Uncertainty and Sensitivity Evaluation) programme. The purpose of BEMUSE is the comparison of best-estimate calculations, followed by the comparison of uncertainty and sensitivity analyses for a Large Break Loss of Coolant Accident (LB-LOCA). The first part of the programme is devoted to the study of the LOFT L2-5 experiment. After a general presentation of the programme, which implies more than ten participants, this paper describes the qualification process and the results of the best-estimate calculations. The results are significantly less dispersed than those of the ISP-13, concerning already LOFT L2-5 more than 20 years ago. Then, it presents extensively the methods and the results of uncertainty and sensitivity analyses. All the participants, apart from the University of Pisa with the CIAU method, use a fully probabilistic approach, based on Wilks’ formula. However, differences appear for the choice of the uncertain input parameters to be considered and for their associated range of variation. Sensitivity analysis is performed with regression techniques, and the results are also compared. As a conclusion, main lessons learnt from BEMUSE and recommendations are presented.
The BEMUSE Programme: Results of the first part concerning the LOFT L2-5 test
D'AURIA, FRANCESCO SAVERIO;
2006-01-01
Abstract
This paper is aimed at describing results of the first part of the BEMUSE (Best Estimate Methods – Uncertainty and Sensitivity Evaluation) programme. The purpose of BEMUSE is the comparison of best-estimate calculations, followed by the comparison of uncertainty and sensitivity analyses for a Large Break Loss of Coolant Accident (LB-LOCA). The first part of the programme is devoted to the study of the LOFT L2-5 experiment. After a general presentation of the programme, which implies more than ten participants, this paper describes the qualification process and the results of the best-estimate calculations. The results are significantly less dispersed than those of the ISP-13, concerning already LOFT L2-5 more than 20 years ago. Then, it presents extensively the methods and the results of uncertainty and sensitivity analyses. All the participants, apart from the University of Pisa with the CIAU method, use a fully probabilistic approach, based on Wilks’ formula. However, differences appear for the choice of the uncertain input parameters to be considered and for their associated range of variation. Sensitivity analysis is performed with regression techniques, and the results are also compared. As a conclusion, main lessons learnt from BEMUSE and recommendations are presented.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
