Research On Fast Reactor Burnup Calculation By Using CRAM Method

With the introduction of the fourth-generation advanced nuclear energy system,fast reactors occupy a place in the fourth-generation advanced nuclear energy system due to its inherent safety,fertile nuclear fuel,and the advantages of long-lived nuclides in transmutable nuclear waste.Fast reactor research has also become an important direction.On the basis of the existing fast reactor three-dimensional hexagonal nodal method program HEXA3D,this thesis first aims at the problem of missing radial leakage term,and improves the high-order fitting of axial flux to further improve the accuracy of diffusion calculation.In addition,HEXA3D's three-dimensional one-sixth core modeling is supplemented with three-dimensional full-reactor modeling functions,which extends the application range of the program to fast reactor calculations that can perform asymmetric modeling.Finally,the fast reactor burnup calculation module FCAC is coupled with the core diffusion calculation module HEX A3 D to realize the continuous calculation of the long burnup step of the fast reactor.In the burnup calculation module,based on the burnup characteristics of fast reactors,first,ignoring the lower cross-section reaction chain in the fast reactor,selecting the burnup chain suitable for fast reactors and making the burnup input including the burnup chain and fission products card.After that,the burnup matrix is calculated according to the flux distribution of each node calculated by the steady state of the reactor core and the microscopic cross-section of each nuclide in the node,and the Chebyshev rational approximation method,that is,the CRAM method,is used to process the burnup matrix.On the approximate order,a 14-order PFD scheme is selected for calculation,and the fuel consumption step length is selected as 100 days.For the correction method of the burnup result,the classic prediction correction method based on nuclear density is selected to improve the burnup calculation result.On this basis,the international lead-based fast reactor RBEC-M benchmark questions were verified.Use NJOY to process the 33 groups of question-independent MATXS cross-sections obtained from the evaluation nuclear database of ENDF/B-?.1.and read and process the MATXS cross-sections generated by NJOY through the cross-section processing software MGGC independently developed by the laboratory.Five kinds of lumped pseudo-fission products are used,and 194 kinds of fission products are considered for each lumped pseudo-fission product,and finally the ISOTXS cross section related to the problem is obtained.The calculation result is compared with the DIF3D calculation result using the same cross-section library.It has been verified that the full-stack calculation function can be implemented correctly.The RBEC-M benchmark problem is only 20pcm away from the one-sixth modeling result under the same cross-section library condition,which is compared with the DIF3D calculation The results are within 200pcm,the maximum deviation of the core power distribution is within 5%,and the trend of the fuel consumption calculation results within 900 days is consistent with the trend calculated by REBUS.It can be concluded that the self-developed fast reactor core burnup program FCAC coupled with the fast reactor hexagonal nodal method coupled with the fast reactor hexagonal nodal method for diffusion calculation can output the fast reactor burnup results under long burnup steps.,It has the advantages of high calculation accuracy and wide application range.In the future,the fuel consumption module can be improved to adapt to more working conditions.