MCBMPI-A parallel Monte Carlo burnup code based on message passing interface
•A domain decomposition parallel Monte Carlo burnup code based on MPI is developed.•The speedup of depletion calculation part is better than transport calculation part.•MCBMPI can solve 3D neutron depletion problem faster with MPI parallel mode. To make whole reactor transport depletion calculation...
Gespeichert in:
Veröffentlicht in: | Annals of nuclear energy 2020-06, Vol.141, p.107343, Article 107343 |
---|---|
Hauptverfasser: | , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •A domain decomposition parallel Monte Carlo burnup code based on MPI is developed.•The speedup of depletion calculation part is better than transport calculation part.•MCBMPI can solve 3D neutron depletion problem faster with MPI parallel mode.
To make whole reactor transport depletion calculation faster, MCBMPI, a parallel Monte Carlo burnup code based on Message Passing Interface (MPI) is developed. The main feature of this version of code is that domain decomposition method is used in depletion parallel calculation, which makes the whole neutron transport depletion calculation in parallel mode efficiently. Two depletion benchmarks are calculated to verify MCBMPI. The first one is Westinghouse high burnup pin cell problem. The results show that MCBMPI agree well with both MCODE and CASMO-4. And the second benchmark is VERA depletion benchmark 2C. This multi-cell fuel assembly problem is chosen to verify the parallel efficiency of MCBMPI. The results show that the speedup of depletion calculation part is better than that of transport calculation part. The verification results show that MCBMPI is satisfactory for depletion calculation and can do the calculation faster with MPI parallel calculation. |
---|---|
ISSN: | 0306-4549 1873-2100 |
DOI: | 10.1016/j.anucene.2020.107343 |