TY - GEN
T1 - Modeling the MSRR using MPACT
AU - Herring, Nicholas
AU - Collins, Benjamin
AU - Clarno, Kevin
N1 - Publisher Copyright:
© 2024 AMERICAN NUCLEAR SOCIETY. All rights reserved.
PY - 2024
Y1 - 2024
N2 - A high-fidelity digital twin is being created for the Molten Salt Research Reactor (MSRR) project using the Virtual Environment for Reactor Applications (VERA). MPACT is the primary transport solver for the VERA suite and this work investigates modeling the MSRR (from the reported preliminary safety analysis report (PSAR) design) using MPACT. Comparisons are made against benchmark continuous energy Shift simulations for each model investigated. Modeling of the 2D core mid-plane indicates good agreement between 252 energy group calculations in MPACT with eigenvalue errors of less than 450 pcm and quarter pin power errors of less than 1%. 69 energy group eigenvalue results for the 2D core are less accurate, with errors of up to 617 pcm, but the quarter pin power errors remain low, with less than 1% maximum error. 3D results show differences between 69 energy group MPACT calculations and the Shift benchmark of less than 376 pcm, while the 252 energy group calculation matches Shift within 223 pcm. Control rod worths computed by MPACT agree with Shift within 100 pcm for two different control rod withdrawal lengths tested. Good agreement on quarter pin powers carries through to the 3D core model, with errors universally under 1.2%. A brief analysis of computational costs shows that the 252 group calculations are roughly three times as expensive as the 69 group calculations, both in terms of memory and in terms of runtime.
AB - A high-fidelity digital twin is being created for the Molten Salt Research Reactor (MSRR) project using the Virtual Environment for Reactor Applications (VERA). MPACT is the primary transport solver for the VERA suite and this work investigates modeling the MSRR (from the reported preliminary safety analysis report (PSAR) design) using MPACT. Comparisons are made against benchmark continuous energy Shift simulations for each model investigated. Modeling of the 2D core mid-plane indicates good agreement between 252 energy group calculations in MPACT with eigenvalue errors of less than 450 pcm and quarter pin power errors of less than 1%. 69 energy group eigenvalue results for the 2D core are less accurate, with errors of up to 617 pcm, but the quarter pin power errors remain low, with less than 1% maximum error. 3D results show differences between 69 energy group MPACT calculations and the Shift benchmark of less than 376 pcm, while the 252 energy group calculation matches Shift within 223 pcm. Control rod worths computed by MPACT agree with Shift within 100 pcm for two different control rod withdrawal lengths tested. Good agreement on quarter pin powers carries through to the 3D core model, with errors universally under 1.2%. A brief analysis of computational costs shows that the 252 group calculations are roughly three times as expensive as the 69 group calculations, both in terms of memory and in terms of runtime.
KW - MPACT
KW - MSRR
KW - Neutronics
UR - https://www.scopus.com/pages/publications/85202830846
U2 - 10.13182/PHYSOR24-43649
DO - 10.13182/PHYSOR24-43649
M3 - Conference contribution
AN - SCOPUS:85202830846
T3 - Proceedings of the International Conference on Physics of Reactors, PHYSOR 2024
SP - 2088
EP - 2097
BT - Proceedings of the International Conference on Physics of Reactors, PHYSOR 2024
PB - American Nuclear Society
T2 - 2024 International Conference on Physics of Reactors, PHYSOR 2024
Y2 - 21 April 2024 through 24 April 2024
ER -