Quadratic axial expansion function with sub-plane acceleration scheme for the high-fidelity transport code PROTEUS-MOC
•Quadratic axial expansion functions implemented in the PROTEUS-MOC transport code.•Sub-plane acceleration scheme for quadratic axial expansion functions of PROTEUS-MOC.•Higher order axial expansion functions reduce the memory requirement.•Solution convergence behavior of PROTEUS-MOC with axial mesh...
Gespeichert in:
Veröffentlicht in: | Annals of nuclear energy 2020-12, Vol.148 (C), p.107713, Article 107713 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •Quadratic axial expansion functions implemented in the PROTEUS-MOC transport code.•Sub-plane acceleration scheme for quadratic axial expansion functions of PROTEUS-MOC.•Higher order axial expansion functions reduce the memory requirement.•Solution convergence behavior of PROTEUS-MOC with axial mesh refinement.
PROTEUS-MOC is a pin-resolved high-fidelity transport code that employs the method of characteristics to discretize the radial variables and the discontinuous finite element method to discretize the axial variable. In this study, we extended the hybrid transport solution method to axial quadratic trial functions from the current linear functions. Compared with the linear approximation, the quadratic approximation reduces the memory requirement by 50% by allowing coarser axial meshes, but the coarser axial meshes deteriorate the performance of the CMFD and pCMFD acceleration schemes. To remedy the deteriorated performance of the acceleration schemes, a sub-plane acceleration method was implemented. The sub-plane acceleration scheme can improve the computational performance of the quadratic approximation in conjunction with the GMRES method. It was also observed that for a fixed radial mesh configuration, the flat leakage source approximation error increases with axial mesh refinement while the truncation error of the finite functional expansion decreases. |
---|---|
ISSN: | 0306-4549 1873-2100 |
DOI: | 10.1016/j.anucene.2020.107713 |