图卷积网络与Stacking集成学习相结合的堆芯自给能中子探测器故障识别方法

doi:10.1007/s12204-023-2684-x

摘要/Abstract

摘要： 自给能中子探测器(self-powered neutron detector，SPND)作为监测反应堆运行状况和安全裕度的重要传感设备，其健康状态直接影响堆芯安全运行。为实现SPND故障识别的准确性和可靠性，提出一种图卷积网络与Stacking集成学习相结合的SPND故障识别方法。利用图卷积网络(graph convolutional network，GCN)抽取堆内不同位置SPND之间的空间邻域信息，实现SPND电流信号的非线性拟合以获得残差；在此基础上，建立多个机器学习算法融合的Stacking集成学习模型以充分提取残差序列的时变动态特征，实现对SPND正常、漂移、偏置、精度下降和完全失效5种运行状态的识别。算例分析表明，相较于单一模型，GCN-Stacking模型具有融合各基学习器多样化和差异化的优势，使得故障识别效果更稳定与可靠，且在反应堆不同功率下，依然具有较高的故障识别准确度。

关键词: 自给能中子探测器, 图卷积网络, Stacking集成学习, 故障识别

Abstract: Self-powered neutron detectors (SPNDs) play a critical role in monitoring the safety margins and overall health of reactors, directly affecting safe operation within the reactor. In this work, a novel fault identification method based on graph convolutional networks (GCN) and Stacking ensemble learning is proposed for SPNDs. The GCN is employed to extract the spatial neighborhood information of SPNDs at different positions, and residuals are obtained by nonlinear fitting of SPND signals. In order to completely extract the time-varying features from residual sequences, the Stacking fusion model, integrated with various algorithms, is developed and enables the identification of five conditions for SPNDs: normal, drift, bias, precision degradation, and complete failure. The results demonstrate that the integration of diverse base-learners in the GCN-Stacking model exhibits advantages over a single model as well as enhances the stability and reliability in fault identification. Additionally, the GCN-Stacking model maintains higher accuracy in identifying faults at different reactor power levels.

中图分类号:

TM623.7

. 图卷积网络与Stacking集成学习相结合的堆芯自给能中子探测器故障识别方法[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(5): 1018-1027.

LIN Weiqing, LU Yanzhen, MIAO Xiren, QIU Xinghua. Fault Identification Method for In-Core Self-Powered Neutron Detectors Combining Graph Convolutional Network and Stacking Ensemble Learning[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(5): 1018-1027.

参考文献

[1] SANG Y D, DENG B J, ZHANG Q M, et al. Development and verification of a simulation toolkit for Self-Powered Neutron Detector [J]. Annals of Nuclear Energy, 2021, 150: 107784.

[2] PENG X J, LI Q, WANG K. Fault detection and isolation for self powered neutron detectors based on Principal Component Analysis [J]. Annals of Nuclear Energy, 2015, 85: 213-219.

[3] XU Y, CAI Y Z, SONG L. Review of research on condition assessment of nuclear power plant equipment based on data-driven [J]. Journal of Shanghai Jiao Tong University, 2022, 56(3): 267-278 (in Chinese).

[4] CHEN Y S, YANG Y H, LIN M, et al. Fault diagnosis technology of nuclear power plant based on deep learning neural network [J]. Journal of Shanghai Jiao Tong University, 2018, 52(S1): 58-61 (in Chinese).

[5] LI W, PENG M J, WANG Q Z. Improved PCA method for sensor fault detection and isolation in a nuclear power plant [J]. Nuclear Engineering and Technology, 2019, 51(1): 146-154.

[6] NAIMI A, DENG J M, SHIMJITH S R, et al. Fault detection and isolation of a pressurized water reactor based on neural network and K-nearest neighbor [J]. IEEE Access, 2022, 10: 17113-17121.

[7] LI W, PENG M J, WANG Q Z. Fault detectability analysis in PCA method during condition monitoring of sensors in a nuclear power plant [J]. Annals of Nuclear Energy, 2018, 119: 342-351.

[8] YU Y E, PENG M J, WANG H, et al. Improved PCA model for multiple fault detection, isolation and reconstruction of sensors in nuclear power plant [J]. Annals of Nuclear Energy, 2020, 148: 107662.

[9] YELLAPU V S, TIWARI A P, DEGWEKER S B. Application of data reconciliation for fault detection and isolation of in-core self-powered neutron detectors using iterative principal component test [J]. Progress in Nuclear Energy, 2017, 100: 326-343.

[10] RAO N S V, GREULICH C, RAMUHALLI P, et al. Estimation of sensor measurement errors in reactor coolant systems using multi-sensor fusion [J]. Nuclear Engineering and Design, 2021, 375: 111024.

[11] ZHANG Q M, DENG B J, LIU X X, et al. Deconvolution-based real-time neutron flux reconstruction for Self-Powered Neutron Detector [J]. Nuclear Engineering and Design, 2018, 326: 261-267.

[12] LIN W Q, MIAO X R, CHEN J, et al. Forecasting thermal parameters for ultra-high voltage transformers using long- and short-term time-series network with conditional mutual information [J]. IET Electric Power Applications, 2022, 16(5): 548-564.

[13] LIU L B, CHEN J W, WU H F, et al. Physical-virtual collaboration modeling for intra- and inter-station metro ridership prediction [J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(4): 3377-3391.

[14] LI T F, ZHOU Z, LI S N, et al. The emerging graph neural networks for intelligent fault diagnostics and prognostics: A guideline and a benchmark study [J]. Mechanical Systems and Signal Processing, 2022, 168: 108653.

[15] PARI R, SANDHYA M, SANKAR S. A multitier stacked ensemble algorithm for improving classification accuracy [J]. Computing in Science & Engineering, 2020, 22(4): 74-85.

[16] GANAIE M A, HU M H, MALIK A K, et al. Ensemble deep learning: A review [J]. Engineering Applications of Artificial Intelligence, 2022, 115: 105151.

[17] LI H J, ZHU J H, FU X F, et al. Ultra-short-term load forecasting of electric vehicle charging stations based on ensemble learning [J]. Journal of Shanghai Jiao Tong University, 2022, 56(8): 1004-1013 (in Chinese).

[18] DONG X B, YU Z W, CAO W M, et al. A survey on ensemble learning [J]. Frontiers of Computer Science, 2020, 14(2): 241-258.

[19] HOU H, LIU C, WANG Q, et al. Load forecasting combining phase space reconstruction and stacking ensemble learning [J]. IEEE Transactions on Industry Applications, 2023, 59(2): 2296-2304.

[20] LUO Z H, FANG C Y, LIU C L, et al. Method for cleaning abnormal data of wind turbine power curve based on density clustering and boundary extraction [J]. IEEE Transactions on Sustainable Energy, 2022, 13(2): 1147-1159.

[21] ZHOU Y, CAO L Z, HE Q M, et al. A coupled deterministic and monte-carlo method for modeling and simulation of self-powered neutron detector [J]. IEEE Transactions on Nuclear Science, 2022, 69(10): 2118-2128.

[22] CAO L Z, LI Z, WU H C. Numerical simulation and sensitivity study of the rhodium self-powered neutron detector used in PWR [J]. IEEE Transactions on Nuclear Science, 2019, 66(4): 742-751.

[23] CHEN L Y, HONG D J, HE X, et al. Distributed photovoltaic real-time output estimation based on graph convolutional networks [J]. Journal of Shanghai Jiaotong University (Science), 2022: 1-7.

[24] SHI J Q, LI C X, YAN X H. Artificial intelligence for load forecasting: A stacking learning approach based on ensemble diversity regularization [J]. Energy, 2023, 262: 125295.

[25] SHI J Q, ZHANG J H. Load forecasting based on multi-model by stacking ensemble learning [J]. Proceedings of the CSEE, 2019, 39(14): 4032-4042 (in Chinese).

[26] LI W, PENG M J, WANG Q Z. Fault identification in PCA method during sensor condition monitoring in a nuclear power plant [J]. Annals of Nuclear Energy, 2018, 121: 135-145.

[27] CHEN J, LU Y Z, JIANG H, et al. Anomaly detection of core self-powered neutron detector based on twin model [J]. Nuclear Power Engineering, 2023, 44(3): 210-216 (in Chinese).

[28] ZHANG Z L, ZHANG Z N, EIKEVIK T M, et al. Ventilation system heating demand forecasting based on long short-term memory network [J]. Journal of Shanghai Jiao Tong University (Science), 2021, 26(2): 129-137.