Three-dimensional relative power distribution is an important physical quantity for characterizing the neutronic state of a reactor core and the spatial uniformity of core power.However, the high-fidelity calculation of three-dimensional relative power distribution heavily relies on large-scale numerical simulations and is computationally expensive, which makes it difficult to satisfy the demands of rapid evaluation, repeated-query tasks and uncertainty quantification in practical applications.To address this issue, a nonlinear reduced order method is developed for the reconstruction and uncertainty characterization of relative power fields by combining the truncated linear modal representation with the variational autoencoder.In the offline phase of the method, high-fidelity snapshot fields are first projected onto a truncated modal space by proper orthogonal decomposition (POD), to obtain the low-dimensional modal coefficients that retain the dominant physical structures of the original power field.Then, a variational autoencoder is employed to learn the nonlinear latent distribution of the modal coefficients, such that the high-dimensional relative power field can be represented and reconstructed in a more compact latent space.On this basis, a parameter-to-latent prediction network is further trained to establish the mapping from physical parameters to latent variables, and the reconstructed power field is obtained through the decoder together with the truncated modal basis.For the online application of the method, by inputting the parameters of physical model, the distribution of latent variables is first obtained and then sampled in the latent space.The method is able to propagate uncertainty from the latent representation to the reconstructed power field, thereby providing uncertainty estimates and confidence intervals for the prediction results.Simulation experiments are carried out for the steady-state three-dimensional relative power distribution of the HPR1000 reactor core of Hualong One, where high-fidelity samples are generated by CORCA-3D and 5% Gaussian noise is added to test the robustness of the method.It is shown that, in comparison with traditional linear reduced order modelling and neural-network-based methods, the method can maintain high reconstruction accuracy with much lower latent-space dimension, the relative error can be controlled close to 1% in the representative settings, while the uncertainty estimates are generally consistent with the local prediction quality and can reflect the reliability of the reconstructed results by providing the corresponding confidence intervals.These results demonstrate that the method provides an effective approach for rapid reactor-core power restruction and uncertainty quantification.
LiX Y, LiuQ W, LiQ, et al.177 core nuclear design for HPR1000 [J].Nuclear Power Engineering, 2019, 40(S1): 8-12.
[2]
AnP, MaY Q, XiaoP, et al.Development and validation of reactor nuclear design code CORCA-3D [J].Nucl Eng Technol, 2019, 51(7): 1721-1728.
[3]
BerkoozG, HolmesP, LumleyJ L.The proper orthogonal decomposition in the analysis of turbulent flows [J].Annu Rev Fluid Mech, 1993, 25: 539-575.
[4]
BennerP, GugercinS, WillcoxK.A survey of projection-based model reduction methods for parametric dynamical systems [J].SIAM Rev, 2015, 57(4): 483-531.
[5]
ZhangQ Y, HuQ X, WangH, et al.A fast method for Darcy-Stokes equation combing model order reduction and machine learning [J].J Sichuan Univ (Nat Sci Ed), 2025, 62(3): 584-590.
BerahmandK, DaneshfarF, SalehiE S, et al.Autoencoders and their applications in machine learning: A survey [J].Artif Intell Rev, 2024, 57(2): 28.
[8]
BrivioS, FrescaS, FrancoN R, et al.Error estimates for POD-DL-ROMs: A deep learning framework for reduced order modeling of nonlinear parametrized PDEs enhanced by proper orthogonal decomposition [J].Adv Comput Math, 2024, 50(3): 33.
[9]
FrancoN R, FraulinD, ManzoniA, et al.On the latent dimension of deep autoencoders for reduced order modeling of PDEs parametrized by random fields [J].Adv Comput Math, 2024, 50(5): 96.
[10]
LeeK, CarlbergK T.Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders [J].J Comput Phys, 2020, 404: 108973.
[11]
ZhangJ C, ZhouY F, ZhangQ, WangX, ZhaoQ.Surrogate model of predicting eigenvalue and power distribution by convolutional neural network [J].Front Energy Res, 2022, 10: 851231.
[12]
YangJ, SuiX, HuangY P, et al.Assessment of reactor flow field prediction based on deep learning and model reduction [J].Ann Nucl Energy, 2022, 179: 109367.
[13]
GongH L, ChengS B, ChenZ, et al.Data-enabled physics-informed machine learning for reduced-order modeling digital twin: Application to nuclear reactor physics [J].Nucl Sci Eng, 2022, 196(6): 668-693.
[14]
YangY, GongH L, HeQ L, et al.On the uncertainty analysis of the data-enabled physics-informed neural network for solving neutron diffusion eigenvalue problem [J].Nucl Sci Eng, 2024, 198(5): 1075-1096.
[15]
GongH L, ChengS B, ChenZ, et al.An efficient digital twin based on machine learning SVD autoencoder and generalised latent assimilation for nuclear reactor physics [J].Ann Nucl Energy, 2022, 179: 109431.
[16]
ChengN J, MalikO A, DeS, et al.Bi-fidelity variational auto-encoder for uncertainty quantification [J].Comput Method Appl M, 2024, 421: 116793.
[17]
ShinH, ChoiM.Physics-informed variational inference for uncertainty quantification of stochastic differential equations [J].J Comput Phys, 2023, 487: 112183.