Bayesian estimation for covariance between cross-section and errors of experiment and calculation

• Published in: Annals of nuclear energy Vol. 163; p. 108589
• Main Authors: Takeda, Satoshi, Sugihara, Hideyuki, Kitada, Takanori
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-12-2021
• Subjects: Bayesian theory; Calculation error; Covariance; Cross-section adjustment; Experimental error; Cross-section adjustment; Bayesian theory; Covariance; Calculation error; Experimental error
• ISSN: 0306-4549; 1873-2100
• DOI: 10.1016/j.anucene.2021.108589

•The adjustment method is derived for XS and errors of experiment and calculation.•The uncertainty is reduced by the adjustment of covariances between these parameters.•The adjustment of the correlation is not required if it is individually adjusted.
The covariance between sensitive cross-sections is adjusted in the cross-section adjustment method even if these cross-sections are originally independent of each other. On the other hand, the experimental error and calculation error are also sensitive to the results, thus the covariance among not only cross-sections but also errors of experiment and calculation can be adjusted. However, the derivation and the impact of the covariance are not well discussed. Therefore, in the present study, the adjustment method is derived for the whole of the cross-section and errors of experiment and calculation. The result of analysis using light-water moderated uranium benchmark shows the covariance between cross-section and errors of experiment and calculation reduces the uncertainty. There might be correlations between the cross-section and the errors of experiment and calculation, however, the adjustment of correlation between them had better be carefully discussed since the correlation is thought to be very weak in practical.