Comparative adjustment in the fast energy range of recently released actinide data using the AGLLS/APIA/SEBBA methodology determining suited assimilation databases

• Nuclear data adjustment. • Application to fast, hard spectrum systems. • Adjusting JENDL-5, ENDF/B-VIII.1 and JEFF-3.3 data. • Overall performance of the posterior data comparable. The performance of recent nuclear data libraries, viz. JENDL-5, ENDF/B-VIII.1 and JEFF-3.3, is compared in terms of adjusted fast data for 235 U, 238 U, 239 Pu and 240 Pu based upon the use of the AGLLS/APIA/SEBBA package. Thereby, AGLLS stands for Asymptotic Generalized Linear Least-Squares, APIA for Asymptotic Progressing nuclear data Incremental Adjustment; SEBBA, the Selection of Experiment-Based Benchmarks Algorithm, refers to a novel method choosing assimilation database parameters from a large pool of dedicated experiment-based benchmarks in a systematic manner. It is shown that the results obtained by means of the adjusted libraries including computed effective multiplication factors ( k eff ) and spectral indices at core center are well comparable, leading for k eff to mean values of χ 2 just slightly larger than 1 and variances of the individual χ 2 s around 3. This is certainly a promising outcome. Despite consistence and these lower values compared to the prior situation, some of the adjusted nuclear data such as the inelastic scattering cross-section of 238 U are more discrepant than in the unperturbed libraries pointing to compensation effects among data libraries. The ENDF/B-VIII.1 library giving among the three unperturbed libraries a better performance in terms of lower mean χ 2 and associated variance, cannot be further adjusted in a meaningful way by reducing χ 2 to a value < 1. This seems substantiating the need of considering additional benchmark data stemming from new experiments and to improve covariance data, with the objective of converging to a unique set of improved data, i.e., fast data for 235 U, 238 U, 239 Pu and 240 Pu in this study.