基于相似性特征的棒束组件域热工水力参数预测

钱浩; 陈广亮; 孙大彬; 李锦超; 殷新立; 张历轩; 张宇航; 李瑞

doi:10.13832/j.jnpe.2025.S1.0026

基于相似性特征的棒束组件域热工水力参数预测

doi: 10.13832/j.jnpe.2025.S1.0026

1.
哈尔滨工程大学核科学与技术学院，哈尔滨，150001
2.
武汉第二船舶设计研究所，武汉，430205

基金项目: 中核领创项目（CNNC-LCKY-2024-053）

详细信息

作者简介:
钱　浩（1998—），男，博士研究生，现主要从事反应堆堆芯热工水力计算分析的研究，E-mail: haoqian@hrbeu.edu.cn

通讯作者:
陈广亮，E-mail: chenguangliang@hrbeu.edu.cn

中图分类号: TL334
计量
- 文章访问数: 4
- HTML全文浏览量: 1
- PDF下载量: 3
- 被引次数: 0
出版历程
- 收稿日期: 2025-03-01
- 修回日期: 2025-05-12
- 刊出日期: 2025-07-09

Prediction of Thermal-Hydraulic Parameters in Rod Bundle Assembly Domain Based on Similarity Features

1.
College of Nuclear Science and Technology, Harbin Engineering University, Harbin, 150001, China
2.
Wuhan Second Ship Design and Research Institute, Wuhan, 430205, China

摘要

摘要: 为准确预测高雷诺数、复杂结构下堆芯流域的热工水力参数，并提高神经网络的预测精度，以快速获悉堆芯热工水力状态，本研究提出了一种新的辅助预测方法。通过对不同工况下棒束通道精细化计算结果进行分析，确定了棒束组件热工水力宏观参数与精细参数分布之间的相似性规律，并以此规律构建神经网络的输入信息，以精确预测温度、压力及速度参数。研究结果表明，本文所构建的代理模型在宏观参数测试数据上的最大均方误差为7.86×10⁻⁴，最小均方误差为1.39×10⁻⁴；在精细参数测试数据上的最大均方误差为9.39×10⁻³，最小均方误差为5.20×10⁻⁴，表明该模型能够准确预测堆芯的热工水力状态。此外，该代理模型在0.504 s内获取堆芯精细化热工水力参数场，数据获取效率较传统方法提升了1149倍，可为构建反应堆堆芯数字孪生体提供有效技术支持。
- 棒束通道 /
- 热工水力 /
- 机器学习 /
- 相似性
Abstract: To accurately predict the thermal-hydraulic parameters of reactor core flow domains under high Reynolds numbers and complex geometries, and to enhance the predictive accuracy of neural networks for rapid assessment of core thermal-hydraulic conditions, this study proposes a novel auxiliary prediction method. An analysis of detailed simulation results of rod bundle channels under various operating conditions identified a similarity pattern between the macroscopic thermal-hydraulic parameters of the rod bundle assemblies and the distribution of fine-scale parameters. This pattern was then utilized to construct the input features of the neural network, enabling precise predictions of temperature, pressure, and velocity parameters. The results demonstrate that the developed surrogate model achieves a maximum mean squared error (MSE) of 7.86×10⁻⁴ and a minimum MSE of 1.39×10⁻⁴ on macroscopic parameter test data. For fine-scale parameter test data, the maximum and minimum MSE are 9.39×10⁻³ and 5.20×10⁻⁴, respectively, indicating the model’s high accuracy in predicting core thermal-hydraulic conditions. Moreover, the surrogate model obtains fine-scale thermal-hydraulic parameter fields of the core within 0.504 seconds—an efficiency improvement of 1149 times compared to traditional methods. This provides a powerful technical foundation for developing digital twins of nuclear reactor cores.
- Rod bundle channel /
- Thermal-hydraulic /
- Machine learning /
- Similarity

HTML全文

图 1 全连接神经网络架构

Figure 1. Fully Connected Neural Network

下载: 全尺寸图片幻灯片

图 2 计算模型示意图

Figure 2. Schematic Diagram of the Computational Model

下载: 全尺寸图片幻灯片

图 3 宏观参数分布趋势

组件标号见图6。

Figure 3. Macroscopic Parameter Distribution Trend

下载: 全尺寸图片幻灯片

图 4 精细参数分布趋势及其差值（轴向位置2.44 m）

组件标号见图6。

Figure 4. Fine-scale Parameter Distribution Trend and Its Differences（Axial Position 2.44 m）

下载: 全尺寸图片幻灯片

图 5 精细参数前三阶模态图（轴向位置2.46 m）

Figure 5. First Three Order Modal Cloud Maps of Fine-scale Parameters（Axial Position 2.46 m）

下载: 全尺寸图片幻灯片

图 6 归一化功率、流量分布

Figure 6. Normalized Power and Flow Distribution

下载: 全尺寸图片幻灯片

图 7 宏观参数预测结果

Figure 7. Macroscopic Parameter Prediction Results

下载: 全尺寸图片幻灯片

图 8 精细参数预测结果

Figure 8. Fine-scale Parameter Prediction Results

下载: 全尺寸图片幻灯片

表 1 组件边界

Table 1. Boundary of the Component

参数	参数值
压力/MPa	15.5
入口温度/℃	292.8
入口平均速度/(m·s⁻¹)	4.06
平均热流密度/(W·m⁻²)	$ 8.447\times {10}^{5}\mathrm{sin}\left(\dfrac{{\text{π}}{\textit{z}}}{3.75}\right) $
z—轴向位置，m。

下载: 导出CSV

表 2 误差表

Table 2. Error Results

参数	宏观参数		精细参数
参数	训练数据	测试数据	训练数据	测试数据
横流速度误差	1.16$ \times $10⁻⁴	7.86$ \times $10⁻⁴	3.40$ \times $10⁻³	5.19$ \times $10⁻³
温度误差	1.18$ \times $10⁻⁴	1.39$ \times $10⁻⁴	4.97$ \times $10⁻³	9.39$ \times $10⁻³
压力误差	5.46$ \times $10⁻⁵	2.52$ \times $10⁻⁴	5.70$ \times $10⁻⁴	5.20$ \times $10⁻⁴

下载: 导出CSV

参考文献(17)

[1]	CONNER M E, HASSAN Y A, DOMINGUEZ-ONTIVEROS E E. Hydraulic benchmark data for PWR mixing vane grid[J]. Nuclear Engineering and Design, 2013, 264: 97-102. doi: 10.1016/j.nucengdes.2012.12.001
[2]	BHATTACHARJEE S, RICCIARDI G, VIAZZO S. Comparative study of the contribution of various PWR spacer grid components to hydrodynamic and wall pressure characteristics[J]. Nuclear Engineering and Design, 2017, 317: 22-43. doi: 10.1016/j.nucengdes.2017.03.011
[3]	CHANG S K, MOON S K, BAEK W P, et al. Phenomenological investigations on the turbulent flow structures in a rod bundle array with mixing devices[J]. Nuclear Engineering and Design, 2008, 238: 600-609. doi: 10.1016/j.nucengdes.2007.02.037
[4]	NGUYEN T, HASSAN Y. Stereoscopic particle image velocimetry measurements of flow in a rod bundle with a spacer grid and mixing vanes at a low Reynolds number[J]. International Journal of Heat and Fluid Flow, 2017, 67: 202-219. doi: 10.1016/j.ijheatfluidflow.2017.08.011
[5]	QU W H, XIONG J B, CHEN S L, et al. High-fidelity PIV measurement of cross flow in 5×5 rod bundle with mixing vane grids[J]. Nuclear Engineering and Design, 2019, 344: 131-143. doi: 10.1016/j.nucengdes.2019.01.021
[6]	YU H, WANG M J, CAI R, et al. Development and validation of boron diffusion model in nuclear reactor core subchannel analysis[J]. Annals of Nuclear Energy, 2019, 130: 208-217. doi: 10.1016/j.anucene.2019.02.046
[7]	JU H R, WANG M J, CHEN C, et al. Numerical study on the turbulent mixing in channel with Large Eddy Simulation (LES) using spectral element method[J]. Nuclear Engineering and Design, 2019, 348: 169-176. doi: 10.1016/j.nucengdes.2019.04.017
[8]	Qi P, Li X, Qiu F, et al. Application of particle image velocimetry measurement technique to study pulsating flow in a rod bundle channel[J]. Experimental Thermal and Fluid Science, 2020, 113: 110047.
[9]	MOORTHI A, SHARMA A K, VELUSAMY K. A review of sub-channel thermal hydraulic codes for nuclear reactor core and future directions[J]. Nuclear Engineering and Design, 2018, 332: 329-344. doi: 10.1016/j.nucengdes.2018.03.012
[10]	WANG M J, WANG Y J, TIAN W X, et al. Recent progress of CFD applications in PWR thermal hydraulics study and future directions[J]. Annals of Nuclear Energy, 2021, 150: 107836. doi: 10.1016/j.anucene.2020.107836
[11]	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
[12]	韩浪,冉旭,单建强,等. 人工神经网络在棒束临界热流密度预测中的应用[J]. 原子能科学技术,2006, 40(3): 257-261. doi: 10.3969/j.issn.1000-6931.2006.03.001
[13]	AYODEJI A, AMIDU M A, OLATUBOSUN S A, et al. Deep learning for safety assessment of nuclear power reactors: reliability, explainability, and research opportunities[J]. Progress in Nuclear Energy, 2022, 151: 104339. doi: 10.1016/j.pnucene.2022.104339
[14]	曾聿赟,刘井泉,杨春振,等. 基于机器学习的小型核反应堆系统状态预测方法[J]. 核动力工程,2018, 39(1): 117-121.
[15]	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 (NURETH-7). La Grange Park: American Nuclear Society, 1995: 3153-3174.
[16]	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
[17]	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(1): 5695809.