Study On The Efficiency Method Of High-Fidelity Depletion Calculation For Numerical Reactor

Pin-resolve depletion calculation is an important part of high-fidelity reactor core physical calculations and an indispensable component of the numerical reactor.High-fidelity depletion calculation analyses the fuel composition changes used to accurately capture the fuel composition changes during the reactor’s lifetime,which play an important role in the analysis of reactor operation and accident conditions.It transforms the complex nuclides relationship into the problem of solving large-scale rigid differential equations,and then uses the numerical solution algorithms to solve the equations.However,the complex conversion relationship between nuclides has brought huge challenges to quickly and accurately solving the burnup equation.Therefore,it is necessary to carry out research on high-fidelity burnup calculation methods,develop efficient and accurate point-depletion code,and realize the internal coupling of neutronic-depletion,which has great significance to the high-fidelity reactor physics analysis.According to the requirements of high-fidelity depletion calculation,the CRAM and QRAM based on the matrix exponent method are selected in this paper.Compared with the general burnup calculation methods,these two methods have the advantages of high calculation accuracy,fast calculation speed,and there is no need to deal with short half-life nuclides separately.An efficient and accurate point-depletion program was developed based on the above methods.The verification results of constant flux neutron irradiation and decay problems show that the point-depletion code developed in this paper can solve the burnup problem quickly and accurately.At present,the self-developed high-fidelity neutron transport calculation program HNET still lacks burnup module,so the solver in this paper is coupled with the neutron transport program HNET.The VERA-1A benchmark question is used to verify the accuracy of the coupling scheme and coupling method,and then compared and analyzed the coupling scheme and coupling method.The numerical results show that the accuracy of the HNET program about depletion questions reaches the same level as that of the international same-type programs.The accuracy of different coupling schemes in the HNET program is not much different,but the efficiency of pin-level coupling scheme is higher.The combination of Prediction-Correction theory and Sub-Step method can improve efficiency while maintaining accuracy well.Using HNET to carry out burnup calculation for the two-dimensional single-pin series of VERA benchmark questions.The Numerical results show that with the calculation conditions of different temperatures,different enrichment degrees,different moderator densities and containing combustible poisons,the error between results of the effective multiplication factor in the burnup process by HNET program and RMC code is within the acceptable range.Conventional fuel assembly and poison assembly in VERA benchmark problem are utilized for burnup calculation.During the burnup process,the error of effective multiplication factor compared with the Serpent and MPACT results are all within 350 pcm.The normalized rod power of the assembly is basically consistent with the power distribution of the Serpent code.According to the above results,HNET already has the ability to calculate actual burnup problems,which has good accuracy and efficiency.The work in this paper further enhances the ability of the HNET code to carry out high-fidelity physical analysis for actual engineering problems.