基于卷积注意力机制的阀门内漏声发射识别方法

黄鑫,屈文忠,肖黎

振动与冲击 ›› 2024, Vol. 43 ›› Issue (9) : 105-114.

PDF(3075 KB)
PDF(3075 KB)
振动与冲击 ›› 2024, Vol. 43 ›› Issue (9) : 105-114.
论文

基于卷积注意力机制的阀门内漏声发射识别方法

  • 黄鑫,屈文忠,肖黎
作者信息 +

Acoustic emission recognition method for valve internal leakage based on convolutional attention mechanism

  • HUANG Xin, QU Wenzhong, XIAO Li
Author information +
文章历史 +

摘要

阀门结构作为核电厂的关键设备部件之一,因长期处于高温高压环境下,其闸板或阀瓣易发生热变形或磨损导致密封不良,进而引发内漏事故。实时在线识别阀门的内漏状态,对提升核电机组热效率、提高阀门可靠性具有重要意义。因实际工业现场的基底噪声极易掩盖阀门内漏的声发射信号,进而造成阀门内漏状态的误判。为实现阀门内漏状态的快速准确识别,搭建了阀门内漏检测试验台架,开发了基于声发射方法的阀门内漏监检测分析系统,将卷积注意力机制引入卷积神经网络中,实现高效快速地识别阀门内漏状态。结果表明:基于阀门内漏的声发射信号频域数据,利用卷积注意力机制神经网络能有效准确地识别阀门内漏状态,在内漏率为26L/h时,识别准确率高达98%,并且具有较好的可靠性和鲁棒性。

Abstract

Because of long-term exposure to high temperature and high-pressure environment, the gate or disc of the valve structure, one of the key equipment components of nuclear power plant, is prone to thermal deformation or wear, resulting in poor sealing, which lead to internal leakage accidents. Real-time on-line identification of internal leakage state of valve is of great significance for improving thermal efficiency and valve reliability of nuclear power units. It’s easy for the acoustic emission signal of valve leakage to be covered by the base noise of the actual industrial site, which result in the misjudgment of valve leakage state. In order to realize quick and accurate identification of valve internal leakage state, in present paper, the valve internal leakage detection test bench is built, the valve internal leakage monitoring and analysis system based on acoustic emission method is developed, and the convolutional block attention module is introduced into the convolutional neural network to realize efficient and rapid identification of valve internal leakage state. The results show that: Based on the acoustic emission signal frequency domain data of the valve leakage, the convolutional block attention module can effectively and accurately identify the valve leakage state. When the internal leakage rate is 26L/h, the recognition accuracy is up to 98%, with good reliability and robustness.

关键词

阀门结构 / 内漏 / 声发射 / 卷积注意力模块 / 卷积神经网络

Key words

Valve structure / Internal leakage / Acoustic emission / Convolutional Block Attention Module / Convolutional Neural Network

引用本文

导出引用
黄鑫,屈文忠,肖黎. 基于卷积注意力机制的阀门内漏声发射识别方法[J]. 振动与冲击, 2024, 43(9): 105-114
HUANG Xin, QU Wenzhong, XIAO Li. Acoustic emission recognition method for valve internal leakage based on convolutional attention mechanism[J]. Journal of Vibration and Shock, 2024, 43(9): 105-114

参考文献

[1] 张征明, 吴莘馨, 何树延. 核安全级阀门的结构力学分析[J]. 阀门, 2004(04):1-4. ZHANG Zhengming, WU Xinxin, HE Shuyan. Structural mechanical analysis of the nuclear safety-related valve[J]. Valve, 2004(04):1-4. [2] 李程, 李强, 罗林, 等. 阀门内漏原因的分析及解决方案[J]. 化工管理, 2021(7):132-133. LI Cheng, LI Qiang, LUO Lin, et al. Cause Analysis and Solution of Valve Internal Leakage[J]. Chemical Enterprise Management, 2021(7):132-133. [3] 马咏, 邱宪苗, 丁兆建, 等. 声发射在核电厂二回路阀门内漏检测中的应用[J]. 科技创新导报, 2021, 18(15):45-47. MA Yong, QIU Xianmiao, DING Zhaojian, et al. Application of Acoustic Emission Method in Internal Leakage Detection of Secondary Loop Valves in Nuclear Power Plant[J]. Science and Technology Innovation Herald, 2021, 18(15):45-47. [4] 刘晓初, 叶邦彦. 轴承磁力探伤中磁悬液的优化配制及监控[J]. 现代制造工程, 2003(4):57-59. LIU Xiaochu, YE Bangyan. Optimum composition and monitoring studied on bearing magnetic power flaw detection[J]. Modern Manufacturing Engineering, 2003(4):57-59. [5] 黑强虎, 高继法, 刘继成, 等. 室内管道分段试压的必要性[J]. 城市住宅, 2017, 24(8):85-87. HEI Qianghu, GAO,Jifa, LIU Jicheng, et al. Necessity of Sectional Pressure-test on Indoor Pipeline[J]. Urban Architecture Space, 2017, 24(8):85-87. [6] Li J, Wang L, Xiong H, et al. Research on Characterization of Asphalt Pavement Performance by Acoustic Emission Technology[J]. International Journal of Pavement Research and Technology, 2023, 16:444–473. [7] Dunegan H. Acoustic Emission: A Promissing Technique[Z]. Livermore CA: Lawkence Radiation Laboratory, 1963 [8] Wichmann, H. and D. Phillips, Acoustic Emission Techniques for Locating Internal Leakage of Redundant Components[J]. Journal of Spacecraft and Rockets, 1984, 21(1): 36-40. [9] Quy, T.B. and J.M. Kim. Crack detection and localization in a fluid pipeline based on acoustic emission signals[J]. Mechanical Systems & Signal Processing, 2021, 150: 107254. [10] GB/T 4213-2008, 气动调节阀[S]. GB/T 4213-2008, Pneumatic industrial process control valves[S]. [11] Ye G, Xu K, Wu W. Standard deviation based acoustic emission signal analysis for detecting valve internal leakage[J]. Sensors and Actuators A: Physical, 2018,283:340-347. [12] Bai R, Xu Q, Meng Z, et al. Rolling bearing fault diagnosis based on multi-channel convolution neural network and multi-scale clipping fusion data augmentation[J]. Measurement, 2021,184:109885. [13] Deng F, Ding H, Yang S, et al. An improved deep residual network with multiscale feature fusion for rotating machinery fault diagnosis[J]. Measurement Science and Technology, 2021,32(2):24002. [14] Li Z, Zhang H, Tan D, et al. A novel acoustic emission detection module for leakage recognition in a gas pipeline valve[J]. Process Safety and Environmental Protection, 2017,105:32-40. [15] 朱沈宾,李振林,王西明等.阀门内漏识别及内漏速率量化技术研究[J].振动与冲击,2022,41(04):167-175. ZHU Shenbin, LI Zhenlin, WANG Ximing, et al. A study on valve internal leakage identification and leakage rate quantification[J]. Journal of Vibration and Shock, 2022, 41(04):167-175. [16] Ye G, Xu K, Wu W. Multi-variable classification model for valve internal leakage based on acoustic emission time–frequency domain characteristics and random forest[J]. Review of Scientific Instruments, 2021,92:25108. [17] Sim H Y, Ramli R, Saifizul A. Valve Leakage Analysis in Reciprocating Compressor by Using Acoustic Emission Technique[A]. Springer Proceedings in Physics, 2019, 218: 355-363. [18] Benz A. Use of acoustic emission techniques for detection of discontinuities[J]. Materials Evaluation, 1998, 56(10): 1215-1222. [19] 杨晶, 李录平, 饶洪德, 等. 基于声发射检测的阀门泄漏故障模式诊断技术研究[J]. 动力工程学报, 2013, 33(6): 455-460+483. Yang Jing, Li Luping, Rao Hongde, et al. Diagnosis of Valve Leakage Fault Patterns Based on Acoustic Emission Detection[J]. Journal of Power Engineering, 2013, 33(6): 455-460, 483. [20] Lee J H, Lee M R, Kim J T, et al. A study of the characteristics of the acoustic emission signals for condition monitoring of check valves in nuclear power plants[J]. Nuclear Engineering and Design, 2006, 236(13): 1411-1421. [21] 张瑞程, 王新颖, 胡磊磊, 等. 基于一维卷积神经网络的燃气管道泄漏声发射信号识别[J]. 中国安全生产科学技术, 2021, 17(2): 104-109. ZHANG Ruicheng, WANG Xinying, HU Leilei, et al. Acoustic emission signal identification of gas pipeline leakage based on one-dimensional convolution neural network[J]. Journal of Safety Science and Technology, 2021,17(2):104-109. [22] Haile M A, Zhu E, Hsu C, et al. Deep machine learning for detection of acoustic wave reflections[J]. Structural Health Monitoring, 2020, 19(5): 1340-1350. [23] Woo S, Park J, Lee J Y, et al. CBAM: Convolutional Block Attention Module [C]//2018 European Conference on Computer Vision. Munich: ECCV, 2018.

PDF(3075 KB)

Accesses

Citation

Detail

段落导航
相关文章

/