Deep Learning Solution Technology for Sparse Data of Multi-Channel Flow Field of PWR Rod Bundle

Qian Hao; Chen Guangliang; Liu Dong; Yu Yang; Jiang Hongwei; Yin Xinli; Yang Yucheng

doi:10.13832/j.jnpe.2024.080039

Volume 46 Issue 2

Apr. 2025

Turn off MathJax

Article Contents

Article Navigation > Nuclear Power Engineering > 2025 > 46(2): 81-89

Qian Hao, Chen Guangliang, Liu Dong, Yu Yang, Jiang Hongwei, Yin Xinli, Yang Yucheng. Deep Learning Solution Technology for Sparse Data of Multi-Channel Flow Field of PWR Rod Bundle[J]. Nuclear Power Engineering, 2025, 46(2): 81-89. doi: 10.13832/j.jnpe.2024.080039

Citation:

Qian Hao, Chen Guangliang, Liu Dong, Yu Yang, Jiang Hongwei, Yin Xinli, Yang Yucheng. Deep Learning Solution Technology for Sparse Data of Multi-Channel Flow Field of PWR Rod Bundle[J]. Nuclear Power Engineering, 2025, 46(2): 81-89. doi: 10.13832/j.jnpe.2024.080039

Citation:

Qian Hao, Chen Guangliang, Liu Dong, Yu Yang, Jiang Hongwei, Yin Xinli, Yang Yucheng. Deep Learning Solution Technology for Sparse Data of Multi-Channel Flow Field of PWR Rod Bundle[J]. Nuclear Power Engineering, 2025, 46(2): 81-89. doi: 10.13832/j.jnpe.2024.080039

PDF( 13830 KB)

Deep Learning Solution Technology for Sparse Data of Multi-Channel Flow Field of PWR Rod Bundle

doi: 10.13832/j.jnpe.2024.080039

1.
College of Nuclear Science and Technology, Harbin Engineering University, Harbin, 150001, China
2.
Science and Technology on Reactor System Design Technology Laboratory, Nuclear Power Institute of China, Chengdu, 610213, China

Received Date: 2024-08-15
Accepted Date: 2025-01-17
Rev Recd Date: 2024-10-11

Available Online: 2025-01-15

Publish Date: 2025-04-02

Abstract

Abstract

The Reynolds number can reach 10⁵ in typical reactor core operating conditions, and the coolant flow exhibits significant nonlinearity. An inevitable mismatch between the actual flow boundaries and states and the ideal flow equations can lead to conflicts between the data and the constraints of the governing equations during the solving process. This mutual restriction can cause difficulties in achieving convergence. This paper developed a sparse data-solving method based on deep learning to address this issue. By designing an adaptive mismatch correction scheme, an adaptive adjustment factor is introduced into the governing equations to correct the ideal model dynamically. This approach overcomes the convergence difficulties and accuracy issues caused by the inconsistency between the data and the equations. Based on this technology, the study further explored flow field-solving strategies under small sample data conditions and designed uniform, velocity-gradient-based, and hybrid point distribution strategies. These strategies aim to optimize the spatial distribution of sample points to improve the overall accuracy of the flow field solutions. The results show that among the three strategies, the uniform point distribution strategy provides the most comprehensive coverage of the overall flow field characteristics and achieves the best optimization effect, with an R² value greater than 0.95 and MSE on the order of 10⁻⁴ to 10⁻³. Moreover, even with only 60 small sample points (7.8% of the original data points), the proposed method can still effectively achieve high-accuracy flow field solutions, providing an efficient and highly applicable solution for solving the multi-channel flow field of PWR reactor core rod bundles under sparse data conditions.
- Physics-Informed Neural Network (PINNs),
- Adjustment factor,
- Sparse data,
- Pressurized water reactor,
- Rod bundle multi-channel,
- Deep learning

FullText(HTML)

References(18)

References

[1]	CHEN G L, WANG J J, ZHANG Z J, et al. Distributed-parallel CFD computation for all fuel assemblies in PWR core[J]. Annals of Nuclear Energy, 2020, 141: 107340. doi: 10.1016/j.anucene.2020.107340
[2]	QU W H, XIONG J B. High-fidelity PIV measurements of turbulent flow in reactor pressure vessel assisted by high-precision matched index of refraction technique[J]. Nuclear Engineering and Design, 2024, 420: 112997. doi: 10.1016/j.nucengdes.2024.112997
[3]	QIAN H, CHEN G L, LI L, et al. Development of supporting platform for the fine flow characteristics of reactor core[J]. Nuclear Engineering and Technology, 2024, 56(5): 1687-1697. doi: 10.1016/j.net.2023.12.023
[4]	ZHANG C, JU H R, ZHANG D L, et al. PIV measurement and numerical investigation on flow characteristics of simulated fast reactor fuel subassembly[J]. Nuclear Engineering and Technology, 2020, 52(5): 897-907. doi: 10.1016/j.net.2019.10.013
[5]	HAN B, ZHU X L, YANG B W, et al. Review of the representative development history on rod bundle mixing coefficient used in subchannel analysis code of PWR[J]. Progress in Nuclear Energy, 2024, 170: 105113. doi: 10.1016/j.pnucene.2024.105113
[6]	ZHANG L X, CHEN G L, TIAN Z F, et al. Advanced CFD modeling of rod bundle channels with integrated porous media and momentum source schemes[J]. International Communications in Heat and Mass Transfer, 2024, 155: 107547. doi: 10.1016/j.icheatmasstransfer.2024.107547
[7]	CHEN G L, QIAN H, LI L, Lei L, et al. Design and analysis of RIF scheme to improve the CFD efficiency of rod-type PWR core[J]. Nuclear Engineering and Technology, 2021, 53(10): 3171-3181. doi: 10.1016/j.net.2021.04.008
[8]	张青山. 有限长平板分离再附流动非定常特性的PIV实验研究——基于POD与DMD模态分解的旋涡动力学分析[D]. 上海: 上海交通大学,2015.
[9]	CAO H S, SUN P W, ZHAO L. PCA-SVM method with sliding window for online fault diagnosis of a small pressurized water reactor[J]. Annals of Nuclear Energy, 2022, 171: 109036. doi: 10.1016/j.anucene.2022.109036
[10]	LI W H, LI J G, YAO J F, et al. Mode decomposition of core dynamics transients using higher-order DMD method[J]. Nuclear Engineering and Design, 2024, 427: 113417. doi: 10.1016/j.nucengdes.2024.113417
[11]	DURAISAMY K, IACCARINO G, XIAO H. Turbulence modeling in the age of data[J]. Annual Review of Fluid Mechanics, 2019, 51(1): 357-377. doi: 10.1146/annurev-fluid-010518-040547
[12]	KARNIADAKIS G E, KEVREKIDIS I G, LU L, et al. Physics-informed machine learning[J]. Nature Reviews Physics, 2021, 3(6): 422-440. doi: 10.1038/s42254-021-00314-5
[13]	RAISSI M, PERDIKARIS P, KARNIADAKIS E G. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations[J]. Journal of Computational Physics, 2019, 378: 686-707. doi: 10.1016/j.jcp.2018.10.045
[14]	刘东,罗琦,唐雷,等. 基于PINN深度机器学习技术求解多维中子学扩散方程[J]. 核动力工程,2022, 43(2): 1-8.
[15]	WANG S P, KARNIADAKIS G E. GMC-PINNs: a new general Monte Carlo PINNs method for solving fractional partial differential equations on irregular domains[J]. Computer Methods in Applied Mechanics and Engineering, 2024, 429: 117189. doi: 10.1016/j.cma.2024.117189
[16]	KAROUTA Z, GU C Y, SCHOELIN B. 3-D flow analyses for design of nuclear fuel spacer[C]//Proceedings of the 7th International Meeting on Nuclear Reactor Thermal-hydraulics. La Grange Park: American Nuclear Society, 1995.
[17]	NAVARRO M A, SANTOS A A C. Evaluation of a numeric procedure for flow simulation of a 5 × 5 PWR rod bundle with a mixing vane spacer[J]. Progress in Nuclear Energy, 2011, 53(8): 1190-1196. doi: 10.1016/j.pnucene.2011.08.002
[18]	CHEN G L, ZHANG Z J, TIAN Z F, et al. Challenge analysis and schemes design for the CFD simulation of PWR[J]. Science and Technology of Nuclear Installations, 2017, 2017: 5695809.