NUCLEAR TECHNIQUES, Volume. 48, Issue 4, 040601(2025)
Combined neural network-based transient thermal hydraulic parameter prediction method for fast reactor core
Figures & Tables(19)
Fig. 5. Preliminary screening of SSA parameters for single-step prediction (a) Mass flow rate, (b) Temperature
Fig. 6. Fine screening of SSA parameters for single-step prediction (a) Mass flow rate, (b) Temperature
Fig. 7. Reconstruction order of single-step prediction (a) Mass flow rate, (b) Temperature
Fig. 8. Preliminary screening of SSA parameters for continuous prediction (a) Mass flow rate, (b) Temperature
Fig. 9. Fine screening of SSA parameters for continuous prediction (a) Mass flow rate, (b) Temperature
Fig. 10. Reconstruction order of continuous prediction (a) Mass flow rate, (b) Temperature
Fig. 11. Results of RBF single-step prediction (a) Mass flow rate, (b) Temperature
Fig. 12. Results of EMD-RBF single-step prediction (a) Mass flow rate, (b) Temperature
Fig. 13. Results of EMD-SSA-RBF single-step prediction (a) Mass flow rate, (b) Temperature
Table 1. Main parameters of CEFR
View tableView in ArticleTable 1. Main parameters of CEFR
参数 Parameters 数值Values 功率Power / MW 65 组件内燃料棒数 Number of fuel rods in the assembly 61 活性区高度 Height of active core / mm 450 燃料棒直径 Fuel rod diameter / mm 6.0 包壳厚度 Cladding thickness / mm 0.3 芯块外径/内径
Outer/inner diameter of fuel pellets / mm
5.2/1/6 定位绕丝直径 Wire diameter / mm 0.95 定位绕丝螺距 Wire pitch / mm 100 堆芯进/出口温度
Core inlet/outlet temperature / ℃
360/530
Table 2. Single-step prediction error of the model
View tableView in ArticleTable 2. Single-step prediction error of the model
预测变量
Predictive variables
神经网络方法
Neural network algorithm
平均相对误差
Mean relative error
质量流量
Mass flow
冀南等的文献值The literature value of JI Nan et al. 0.057 8 本文RBF神经网络This paper presents RBF neural network 0.048 8 EMD-RBF组合神经网络EMD-RBF combined neural network 0.028 6 EMD-SSA-RBF组合神经网络EMD-SSA-RBF combined neural network 0.006 5 包壳温度
Temperature
冀南等的文献值The literature value of JI Nan et al. 0.005 1 本文RBF神经网络This paper presents RBF neural network 0.003 8 EMD-RBF组合神经网络EMD-RBF combined neural network 0.002 1 EMD-SSA-RBF组合神经网络EMD-SSA-RBF combined neural network 0.001 5
Table 3. Continuous prediction errors of the model
View tableView in ArticleTable 3. Continuous prediction errors of the model
预测变量
Predictive variables
神经网络方法
Neural network algorithm
平均相对误差
Mean relative error
质量流量
Mass flow
本文RBF神经网络RBF neural network presented in this paper 0.050 2 EMD-SSA-RBF组合神经网络EMD-SSA-RBF combined neural network 0.021 7 包壳温度
Temperature
本文RBF神经网络RBF neural network presented in this paper 0.008 1 EMD-SSA-RBF组合神经网络EMD-SSA-RBF combined neural network 0.004 2
Tools
Get Citation
Copy Citation Text
Ziyan ZHAO, Pengcheng ZHAO, Zijing LIU, Wei LI, Tao YU. Combined neural network-based transient thermal hydraulic parameter prediction method for fast reactor core[J]. NUCLEAR TECHNIQUES, 2025, 48(4): 040601
Paper Information
Category: NUCLEAR ENERGY SCIENCE AND ENGINEERING
Received: Dec. 13, 2023
Accepted: --
Published Online: Jun. 3, 2025
The Author Email: Pengcheng ZHAO (赵鹏程), Tao YU (于涛)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20