采用门控循环神经网络的核工业管道损伤识别方法

PDF(2543 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (24) : 48-58.

论文

采用门控循环神经网络的核工业管道损伤识别方法

蒋琪1，张望2，屈文忠1，肖黎1

作者信息 +

Nuclear industry pipeline damage identification through a gated recurrent neural network

JIANG Qi1，ZHANG Wang2，QU Wenzhong1，XIAO Li1

Author information +

文章历史 +

摘要

超声导波检测技术具有效率高、成本低、检测方便等优点，广泛应用于管道的损伤检测。但超声导波在管道中的传播以及压电传感器灵敏度受温度和压力载荷等环境及工况的影响，严重干扰了损伤信息的提取和识别。由此，本文提出了一种基于粒子群优化-双向门控循环单元-注意力机制模型的机器学习的导波管道损伤识别方法。此模型通过在原始超声导波数据与管道状态之间建立映射关系，并加强特征提取层对损伤特征的识别能力，有效避免环境干扰并准确识别出真实的损伤信号。以某核工业循环水冷却管道试验台架为实验对象，进行温度压力变化工况下的管道损伤识别实验，通过实验和理论分析验证了该模型能有效实现管道损伤识别，且识别准确率优于门控循环网络、长短时记忆、双向门控循环网络等其它数据驱动模型，证实了本文所提方法的有效性和优越性。

Abstract

Ultrasonic guided wave detection technology has the advantages of high efficiency, low cost and convenient detection, and is widely used in pipeline damage detection. However, the propagation of ultrasonic guided wave in the pipeline is affected by the environment such as temperature and load, which seriously interferes with the extraction and recognition of damage information. Therefore, a machine learning pipeline damage identification method based on particle swarm optimization - bidirectional gated recurrent unit - attention mechanism (PSO-BiGRU-Attention) model is proposed in this paper. The model effectively establishes a mapping between raw ultrasonic guided wave data and the state of the pipeline, thereby enhancing the capability of the feature extraction layer to discern damage characteristics, mitigating environmental interferences, and accurately detecting authentic damage signals. Taking the test bench of circulating water cooling pipeline in nuclear industry as the experimental object, the pipeline damage identification experiment under the condition of temperature and load changes was carried out. Through the experimental analysis, it is verified that the model can effectively realize the pipeline damage identification, and the recognition accuracy is better than other models such as recurrent neural network, long short-term memory, bidirectional gated recurrent unit, etc., which proves the effectiveness and superiority of the proposed method in this paper.

导出引用

蒋琪1, 张望2, 屈文忠1, 肖黎1. 采用门控循环神经网络的核工业管道损伤识别方法[J]. 振动与冲击, 2024, 43(24): 48-58

JIANG Qi1, ZHANG Wang2, QU Wenzhong1, XIAO Li1. Nuclear industry pipeline damage identification through a gated recurrent neural network[J]. Journal of Vibration and Shock, 2024, 43(24): 48-58

参考文献

[1] 刘洋, 邓菲, 诸葛霞, 等. 环境参数变化对管道结构健康检测的导波检测影响探究[J]. 仪表技术与传感器, 2018(02): 139-144.
LIU Yang, DENG Fei, ZHUGE Xia, et al. Research on Influence of Environmental Parameters on Guided Wave Detection for Health Detection of Pipeline Structure[J]. Instrument Technique and Sensor, 2018(02): 139-144.
[2] WANG Z, HUANG S, WANG S, et al. Compressed Sensing Method for Health Monitoring of Pipelines Based on Guided Wave Inspection[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(7): 4722-4731.
[3] XU C, YANG Z, CHEN X, et al. A Guided Wave Dispersion Compensation Method Based On Compressed Sensing[J]. Mechanical Systems and Signal Processing, 2018, 103: 89-104.
[4] WANG Y, LI X. Propagation Characteristics and Defect Sensitivity Analysis of Guided Wave From Single Excitation Source in Elbows[J]. IEEE Access, 2019, 7: 75542-75549.
[5] MOUNTASSIR M El, YAACOUBI S, DAHMENE F. Reducing False Alarms in Guided Waves Structural Health Monitoring of Pipelines: Review Synthesis and Debate[J]. International Journal of Pressure Vessels and Piping, 2020, 188: 104210.
[6] GORGIN R, LUO Y, WU Z. Environmental and Operational Conditions Effects On Lamb Wave Based Structural Health Monitoring Systems: A Review[J]. Ultrasonics, 2020, 105: 106114.
[7] EYBPOOSH M, BERGES M, NOH H Y. An Energy-Based Sparse Representation of Ultrasonic Guided-Waves for Online Damage Detection of Pipelines Under Varying Environmental and Operational Conditions[J]. Mechanical Systems and Signal Processing, 2017, 82: 260-278.
[8] 李忠虎, 李昊霖, 李刚, 等. 超声导波管道缺陷检测中的温度补偿方法研究[J]. 内蒙古科技大学学报, 2021, 40(03): 251-255.
LI Zhonghu, LI Haolin, LI Gang, et al. Study on the Influence of Temperature on Nondestructive Testing of Ultrasonic Guided Wave Pipelines [J]. Journal of Inner Mongolia University of Science and Technology, 2021, 40(03): 251-255.
[9] 邓菲, 刘洋, 诸葛霞, 等. 变化环境下的超声导波结构健康监测研究进展[J]. 机械工程学报, 2016, 52(18): 1-7.
DENG Fei, LIU Yang, ZHUGE Xia, et al. Progress on the Research of Ultrasonic Guided Wave Structural Health Monitoring in the Changing Ambient[J]. Journal of Mechanical Engineering, 2016, 52(18): 1-7.
[10] ABBAS S, LI F, ZHU Y, et al. Experimental Investigation of Impact of Environmental Temperature and Optimal Baseline for Thermal Attenuation in Structural Health Monitoring Based On Ultrasonic Guided Waves[J]. Wave Motion, 2020, 93: 102474.
[11] YAN A M, KERSCHEN G, BOE P De, et al. Structural Damage Diagnosis Under Varying Environmental Conditions—Part I: A Linear Analysis[J]. Mechanical Systems and Signal Processing, 2005, 19(4): 847-864.
[12] ROY S, LONKAR K, JANAPATI V, et al. A Novel Physics-Based Temperature Compensation Model for Structural Health Monitoring Using Ultrasonic Guided Waves[J]. Structural Health Monitoring, 2014, 13(3): 321-342.
[13] WANG G, WANG Y, SUN H, et al. A Reference Matching-Based Temperature Compensation Method for Ultrasonic Guided Wave Signals[J]. Sensors, 2019, 19(23): 5174.
[14] WANG Y, WANG G, WU D, et al. An Improved Matching Pursuit-Based Temperature and Load Compensation Method for Ultrasonic Guided Wave Signals[J]. IEEE Access, 2020, 8: 67530-67541.
[15] YE Q, REN Y, QIU L, et al. Accurate Multiple-Site Damage Localization of Aircraft Composite Structures Under Environmental and Operational Conditions[C]. 2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing), 2021, 15-17.
[16] LIU H, ZHANG Y. Deep Learning Based Crack Damage Detection Technique for Thin Plate Structures Using Guided Lamb Wave Signals[J]. Smart Materials and Structures, 2020, 29(1): 15032.
[17] CHEN J X, JIANG D M, ZHANG Y N. A Hierarchical Bidirectional GRU Model with Attention for EEG-Based Emotion Classification[J]. IEEE Access, 2019, 7: 118530-118540.
[18] XIANG L, WANG P, YANG X, et al. Fault Detection of Wind Turbine Based On SCADA Data Analysis Using CNN and LSTM with Attention Mechanism[J]. Measurement, 2021, 175: 109094.
[19] 焦玲玲, 陈捷, 刘连华. 基于CAE和AGRU的滚动轴承退化趋势预测[J]. 振动与冲击, 2023, 42(12): 109-117.
JIAO Lingling, CHEN Jie, LIU Lianhua. Degradation trend prediction of rolling bearings based on CAE and AGRU[J]. Journal of Vibration and Shock, 2023, 42(12): 109-117.
[20] XIA S, ZHOU X, SHI H, et al. A Fault Diagnosis Method Based On Attention Mechanism with Application in Qianlong-2 Autonomous Underwater Vehicle[J]. Ocean Engineering, 2021, 233: 109049.
[21] 石静雯, 侯立群. 基于一维卷积注意力门控循环网络和迁移学习的轴承故障诊断[J]. 振动与冲击, 2023, 42(03): 159-164+173.
SHI Jingwen, HOU Liqun. Bearing fault diagnosis based on 1D CNN attention gated recurrent network and transfer[J]. Journal of Vibration and Shock, 2023, 42(03): 159-164+173.
[22] MITRA S, MUKHOPADHYAY R, CHATTOPADHYAY P. PSO Driven Designing of Robust and Computation Efficient 1D-CNN Architecture for Transmission Line Fault Detection[J]. Expert Systems with Applications, 2022, 210: 118178.
[23] 陈震, 王亚茹, 陈璐, 等. 基于学习因子优化的粒子群算法识别结构损伤[J]. 华北水利水电大学学报(自然科学版), 2022, 43(04): 43-47+75.
CHEN Zhen, WANG Yaru, CHEN Lu, et al. Structural Damage ldentification Based on lmproved Particle Swarm Optimization Algorithm with Learning Factors[J]. Journal of North China University of Water Resources and Electric Power(Natural Science Edition), 2022, 43(04): 43-47+75.
[24] XU L, YANG J, GE M, et al. Three-dimensional fatigue crack quantification using densely connected convolutional network-assisted ultrasonic guided waves[J]. International Journal of Fatigue, 2024, 180: 108094.
[25] RAGHAVAN A, CESNIK C E S. Effects of Elevated Temperature On Guided-wave Structural Health Monitoring[J]. Journal of Intelligent Material Systems and Structures, 2008, 19(12): 1383-1398.
[26] SCALEA F L Di, SALAMONE S. Temperature Effects in Ultrasonic Lamb Wave Structural Health Monitoring Systems[J]. The Journal of the Acoustical Society of America, 2008, 124(1): 161-174.
[27] FAN Z, NAUSHEEN S, PARIA S. Deep learning in fault detection and diagnosis of building HVAC systems: A systematic review with meta analysis[J]. Energy and AI, 2023, 12: 100235.