Application of APEC method in CANDU analysis

Abstract

The CANDU6 (CANada Deuterium Uranium) reactor is a heavy water cooled reactor, four of which are being operated in Korea. Currently, the CANDU6 reactor analysis relies on the Canadian AECL's coarse-mesh FDM code, RFSP. However, the RFSP code is found to be subject to inconsistency issues, this is mainly due to the lack of nodal equivalence. Several attempts have been made to apply the simplified equivalence theory commonly used in PWR core analysis to CANDU core analysis, but they have not been very successful. In this study, the APEC (albedo-corrected parameterized equivalences constants) method is introduced to consider the nodal equivalence theory in the CANDU reactor core calculations. The APEC method, recently developed at KAIST, is a new leakage correction method based on the PWR fuel assembly homogenized two-group cross-sections. In the APEC method, two-group homogenized cross sections can be parameterized as a function of assembly-wise current-to-flux ratio which is the net neutron leakage from the assembly. It was shown that two-group cross section can be corrected very close to the reference value and the nodal accuracy was improved. The CANDU core analysis was performed based on the well-known two-step procedure. The Monte Carlo code Serpent2 was used to perform lattice calculations, including color-set analyses for the APEC method. The APEC function was determined from the least square fitting using color-set calculation results. An in-house nodal expansion method (NEM) code in conjunction with a partial-current based CMFD (p-CMFD) acceleration was used for the full-core calculation. As a result, the cross-sections are corrected significantly which reflects in the nodal accuracy.