Code validation based on the technique of sensitivity and uncertainty analysis

The nuclear data and computationalcodes are basically required for the modeling and simulations of the reactor physics, which should be validated before the applications. The validation procedure must be implemented through comparing the simulated results and the measurements. However, for the newly-designed systems without any available experiment or measurement data, it becomes a big issue to validate the computational codes, which may have been validated within some other applicability area. Therefore, the purpose of this paper is to introduce the methodology for the validation of the reactor-physics codes based on the technique of sensitivity and uncertainty analysis. The sensitivity analysis provides the energy-dependent relative sensitivity coefficients that give the relative change in keff of the reactor system with respect to the relative changes in the cross-section data by isotope, reaction and energy. The uncertainty analysis is used to quantify the uncertainties of the important responses of the reactor system due to the nuclear-data uncertainties. The results of sensitivity and uncertainty analysis, along with the calculated values and measurements of thekeff and corresponding uncertainties of the measurements, are then provided to the Generalized Linear Least-Square Method (GLLSM) to determine the best-estimated "measured'Value for the design system. The GLLSM procedure produces the best-estimated keff value of the newly-designed system, with corresponding relative changes to the cross sections. The best-estimated keff value can be treated as the "measured"value of the design system and can be applied for the code validation. In this paper, the code-validation technique based on the sensitivity and uncertainty analysis is applied to validate our home-developed SARAX code for the China Fast Reactor. From the numerical results, it has been observed that code-validation procedure can provide satisfied results.