Adaptive Coupling of Transport and Depletion Using the Neutronically Weighted Inventory Deviation (NWID) Metric
Authors:
- Abdullah Mohammed AlharbiAAAbdullah Mohammed Alharbi
- Justin Kyle WatsonJWJustin Kyle Watson
- Paul RomanoPRPaul Romano
- Saqr AlshogeathriSASaqr Alshogeathri
Preprint
- 23 June 2026
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
Most depletion engines offer either fully-coupled or fully-independent transport-depletion coupling modes: coupled depletion provides the highest fidelity but becomes impractical for many-interval histories, meanwhile, independent depletion is cheap but loses accuracy when the neutron spectrum changes. This paper introduces adaptive coupling as a new option that applies to both within-interval and across-interval depletion problems, triggering transport recalculation only when the Neutronically Weighted Inventory Deviation (NWID) tracking metric exceeds a user-defined tolerance. NWID quantifies inventory drift weighted by neutronic importance and is computed after each depletion interval at negligible cost. Demonstrations on PWR and SFR pin cells depleted at an irregular and nonlinear schedule to 45 and 131 MWd/kg, respectively, show that end-of-life baseline NWID differs by an order of magnitude between the two systems (145% for the PWR versus 14% for the SFR), mirroring the gap in their fully independent inventory errors and consistent with the spectral rigidity of fast systems. At 1% NWID tolerance for both cases, the PWR recovers 0.85% maximum actinide error, 0.08% maximum fission product error, and 41 pcm k_eff deviation throughout the depletion history all relative to the fully coupled reference. While the SFR achieves 0.09% actinide error, 0.05% fission product error, and a k_eff deviation of 20 pcm. Benchmarking results presented above were achieved while reducing the number of transport solves by 2x and 5x for PWR and SFR, respectively, which directly led to a comparable contribution in runtime reductions.Keywords
This publication has 11 references indexed in Scilit:
- Depletion capabilities in the OpenMC Monte Carlo particle transport codeAnnals of Nuclear Energy, 2021
- ONIX: An open-source depletion codeAnnals of Nuclear Energy, 2021
- Fuel depletion analysis of a small sodium fast reactor with KANEXT and SERPENTAnnals of Nuclear Energy, 2016
- Higher-Order Chebyshev Rational Approximation Method and Application to Burnup EquationsNuclear Science and Engineering, 2016
- Substep methods for burnup calculations with Bateman solutionsAnnals of Nuclear Energy, 2011
- Rational Approximations to the Matrix Exponential in Burnup CalculationsNuclear Science and Engineering, 2011
- Higher order methods for burnup calculations with Bateman solutionsAnnals of Nuclear Energy, 2011
- Comparison of depletion algorithms for large systems of nuclidesAnnals of Nuclear Energy, 2011
- Computing the Matrix Exponential in Burnup CalculationsNuclear Science and Engineering, 2010
- General solution of Bateman equations for nuclear transmutationsAnnals of Nuclear Energy, 2006