A Dataset of Multigroup Relative Covariance Matrix for Average Number of Neutrons Released per Fission in Uncertainty Analysis

Multigroup nuclear reaction cross sections and their associated covariance data are essential input parameters for conducting sensitivity analysis and uncertainty quantification in nuclear reactor neutronics calculations. The average total number of neutrons released per fission propagates its uncertainty into subsequent reactor physics calculations via the covariance matrix, thereby exerting a direct influence on the resulting predictions. This study aims to address the approximation issues inherent in existing calculation models for multigroup covariance matrices. A precise method for calculating the relative covariance matrix of the average total number of neutrons released per fission is proposed, enabling accurate computation of the weighting function on a unionized energy grid. Consequently, a multigroup relative covariance matrix dataset oriented toward uncertainty analysis is established. Based on the ENDF/B-VIII.1 evaluated nuclear data library, this work provides relative covariance matrix data for the total (MT452), prompt (MT456), and delayed (MT455) average total number of neutrons released per fission of the uranium isotopes 233U, 235U, and 238U under three typical energy group structures: the epri-cpm 69-group, lanl 187-group, and xmas lwpc 172-group structures. Numerical results demonstrate that when the user-defined energy group structure aligns with the division of the unionized energy grid, the approximation inherent in the weighting function calculation becomes negligible. Under this condition, the results obtained by the proposed method are consistent with the multigroup relative covariance matrices processed by NJOY2016 within the bounds of numerical precision. For the three typical group structures evaluated, the most significant improvement is observed in the MT452–MT455 relative covariance matrix for 233U, with an average enhancement in computational accuracy of 2.42%. This study effectively resolves the weighting function approximation problem present in existing theoretical methods on the unionized energy grid. The resulting dataset provides critical input parameters for sensitivity and uncertainty analysis in nuclear reactor neutronics calculations and offers a unified data foundation for the improvement, validation, and reproduction of average fission neutron yield covariance processing methodologies.