Implementation and performance study of lpCMFD acceleration method for multi-energy group k-eigenvalue neutron transport problem in hexagonal geometry

Chan, Yimeng; Xiao, Sicong

doi:10.1016/j.anucene.2019.107220

Cited by 5 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Employing the linear prolongation in pCMFD results in a lp-pCMFD scheme, which can also be considered. The lp-CMFD scheme has been employed in deterministic transport problems with various spatial discretisation schemes, which demonstrated improved acceleration and stability (Chan and Xiao, 2019a;Chan and Xiao, 2019b;Chan and Xiao, 2019c;Chan and Xiao, 2020a;Chan and Xiao, 2020b). Following the results in deterministic calculations, lp-CMFD and lp-pCMFD schemes has also been employed in 1D slab MC transport calculations (Chan and Xiao, 2021), as well as in a 2D homogenous slab (Abdullatif and Wang, 2022;Than and Xiao, 2022), where its effectiveness relative to regular CMFD is once again demonstrated.…”

Section: Introductionmentioning

confidence: 99%

lp-CMFD acceleration schemes in multi-energy group 2D Monte Carlo transport

Than

Xiao

2022

Front. Energy Res.

View full text Add to dashboard Cite

The linear prolongation flux update scheme is extended to both regular CMFD acceleration, as well as partial CMFD acceleration in 2D multi energy group Monte Carlo k-eigenvalue neutron transport problems. The acceleration performance of these CMFD variants were investigated in simple 2D slab geometries, first with a monoenergetic case and then with a three group problem on the same geometry based on the monoenergetic cross sections. Flux convergence was determined via an on-the-fly convergence diagnostics developed by Ueki and Brown. It is found that on top of providing better acceleration in general, the linear prolongation scheme is also able to correct for instabilities in the CMFD scheme. Overall, the lp-pCMFD scheme employing a maximum history length is found to have the best performance across the cases presented.

show abstract