TY - JOUR
T1 - Reliability estimation of an advanced nuclear fuel using coupled active learning, multifidelity modeling, and subset simulation
AU - Dhulipala, Somayajulu L.N.
AU - Shields, Michael D.
AU - Chakroborty, Promit
AU - Jiang, Wen
AU - Spencer, Benjamin W.
AU - Hales, Jason D.
AU - Labouré, Vincent M.
AU - Prince, Zachary M.
AU - Bolisetti, Chandrakanth
AU - Che, Yifeng
N1 - Funding Information:
This research is supported through the INL Laboratory Directed Research & Development (LDRD), USA Program under DOE Idaho Operations Office Contract DE-AC07-05ID14517. This research made use of the resources of the High Performance Computing Center at INL, which is supported by the Office of Nuclear Energy of the U.S. DOE and the Nuclear Science User Facilities under Contract No. DE-AC07-05ID14517.
Funding Information:
We thank Peter German at Idaho National Laboratory for his valuable feedback on a draft of this paper. This manuscript has been authored by Battelle Energy Alliance, LLC under Contract No. DE-AC07-05ID14517 with the US Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. This research is supported through the INL Laboratory Directed Research & Development (LDRD), USA Program under DOE Idaho Operations Office Contract DE-AC07-05ID14517. This research made use of the resources of the High Performance Computing Center at INL, which is supported by the Office of Nuclear Energy of the U.S. DOE and the Nuclear Science User Facilities under Contract No. DE-AC07-05ID14517.
Funding Information:
This manuscript has been authored by Battelle Energy Alliance, LLC under Contract No. DE-AC07-05ID14517 with the US Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes.
Publisher Copyright:
© 2022
PY - 2022/10
Y1 - 2022/10
N2 - Tristructural isotropic (TRISO)-coated particle fuel is a robust nuclear fuel and determining its reliability is critical for the success of advanced nuclear technologies. However, TRISO failure probabilities are small and the associated computational models are expensive. We used coupled active learning, multifidelity modeling, and subset simulation to estimate the failure probabilities of TRISO fuels using several 1D and 2D models. With multifidelity modeling, we replaced expensive high-fidelity (HF) model evaluations with information fusion from two low-fidelity (LF) models. For the 1D TRISO models, we considered three multifidelity modeling strategies: only Kriging, Kriging LF prediction plus Kriging correction, and deep neural network (DNN) LF prediction plus Kriging correction. While the results across these multifidelity modeling strategies compared satisfactorily, strategies employing information fusion from two LF models called the HF model least often. Next, for the 2D TRISO model, we considered two multifidelity modeling strategies: DNN LF prediction plus Kriging correction (data-driven) and 1D TRISO LF prediction plus Kriging correction (physics-based). The physics-based strategy, as expected, consistently required the fewest calls to the HF model. However, the data-driven strategy had a lower overall simulation time since the DNN predictions are instantaneous, and the 1D TRISO model requires a non-negligible simulation time.
AB - Tristructural isotropic (TRISO)-coated particle fuel is a robust nuclear fuel and determining its reliability is critical for the success of advanced nuclear technologies. However, TRISO failure probabilities are small and the associated computational models are expensive. We used coupled active learning, multifidelity modeling, and subset simulation to estimate the failure probabilities of TRISO fuels using several 1D and 2D models. With multifidelity modeling, we replaced expensive high-fidelity (HF) model evaluations with information fusion from two low-fidelity (LF) models. For the 1D TRISO models, we considered three multifidelity modeling strategies: only Kriging, Kriging LF prediction plus Kriging correction, and deep neural network (DNN) LF prediction plus Kriging correction. While the results across these multifidelity modeling strategies compared satisfactorily, strategies employing information fusion from two LF models called the HF model least often. Next, for the 2D TRISO model, we considered two multifidelity modeling strategies: DNN LF prediction plus Kriging correction (data-driven) and 1D TRISO LF prediction plus Kriging correction (physics-based). The physics-based strategy, as expected, consistently required the fewest calls to the HF model. However, the data-driven strategy had a lower overall simulation time since the DNN predictions are instantaneous, and the 1D TRISO model requires a non-negligible simulation time.
KW - Active learning
KW - Deep neural networks
KW - Gaussian process
KW - Monte Carlo
KW - Nuclear fuel
KW - Reliability
KW - TRISO fuel
KW - Variance reduction
UR - https://www.scopus.com/pages/publications/85133799684
U2 - 10.1016/j.ress.2022.108693
DO - 10.1016/j.ress.2022.108693
M3 - Article
AN - SCOPUS:85133799684
SN - 0951-8320
VL - 226
SP - 108693
JO - Reliability Engineering and System Safety
JF - Reliability Engineering and System Safety
M1 - 108693
ER -