基于机器学习的在轨航天器泄漏源实时定位方法研究

摘要
图/表
参考文献(17)
相关文章 (15)

全文: PDF (2141 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要本文针对航天器在轨泄漏问题，提出了一种基于机器学习的泄漏源实时定位方法。该方法通过合理特征化泄漏在器壁中激发的弹性波信号，结合有限元仿真技术获得有效的训练数据，设计并实现了一种多层感知机网络模型，从而准确完成泄漏源与传感器间的距离信息估计，同时结合计算弹性波数据时空相关性得到的相对角度信息，可以快速稳定的获得泄漏源的空间位置。本方法仅采用一个布放在器壁上的压电阵列式传感器采集泄漏激发的弹性波数据，结构相对简单。实验结果表明，基于本文设计的多层感知机模型，在1m2实验板范围内该方法对泄漏源与阵列式传感器距离估计准确率为100%，最大定位误差为1.2cm。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	边旭1
	2
	田璧菀1
	靳世久2

关键词 ：弹性波, 泄漏, 定位, 阵列式传感器, 多层感知机

Abstract：Aiming at the problem of spacecraft on-orbit leakage, this paper proposes a real-time localization method based on machine learning. This method reasonably characterizes the elastic wave signal excited by the leak in the wall. Combined with the finite element simulation technology, a multi-layer perceptron (MLP) network model is designed and implemented. Using this model, the distance information between the leak source and the sensor array can be quickly obtained. At the same time, combined with the relative angle information obtained by calculating the spatiotemporal correlation of elastic wave data, the spatial position of the leak source can be obtained quickly and stably. This method uses a piezoelectric array sensor placed on the wall to collect the elastic wave data excited by leakage. The experimental results show that within the range of 1m2 experimental board, based on the multi-layer perceptron model trained in this paper, the method can estimate the distance between the leakage source and the array sensor with an accuracy of 100%, the maximum positioning error is 1.2cm.

Key words： Elastic wave Leakage Location Array sensor MLP

收稿日期: 2022-06-14 出版日期: 2023-08-15

引用本文:

边旭1,2，田璧菀1，靳世久2. 基于机器学习的在轨航天器泄漏源实时定位方法研究[J]. 振动与冲击, 2023, 42(15): 319-324.
BIAN Xu1,2, TIAN Biwan1, JIN Shijiu2. Real time location method for leakage sources in orbiting spacecraft based on machine learning. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(15): 319-324.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I15/319

[1] Wilson WC, Coffey NC, Madaras EI. Leak Detection and Location Technology Assessment for Aerospace Applications [J]. NASA/EM, 2008
[2] Builo S I. Acoustic-Emission Diagnostics of the Effect of Hydrogen on Properties of Materials [J]. Russ J Nondestruct+, 2009, 45 (11): 818-21.
[3] Murvay P S, Silea I. A Survey On Gas Leak Detection and Localization Techniques [J]. Journal of Loss Prevention in the Process Industries, 2012, 25 (6): 966-73.
[4] 马永成, 陈青松. 基于超声的气体泄漏检测与定位技术在载人航天器中的应用[J]. 仪表技术与传感器, 2009, 000(B11): P.341-343.
Ma yong cheng, Chen qing song. Application of Ultrasonic-based Gas Leak Detection and Location Technology in Manned Spacecraft [J] Instrumentation Technology and Sensors, 2009, 000(B11): P.341-343.
[5] Xiao Q, Li J, Sun J, et al. Natural-gas pipeline leak location using variational mode decomposition analysis and cross-time–frequency spectrum[J]. Measurement, 2018, 124: 163-172.
[6] Davoodi S, Mostafapour A. Gas leak locating in steel pipe using wavelet transform and cross-correlation method [J]. International Journal of Advanced Manufacturing Technology, 2014, 70: 1125–1135.
[7] Filho S L M R , Oliveira J A D , Lauro C H , et al. Monitoring of microturning process using acoustic emission signals[J]. Journal of the Brazilian Society of Mechanical Sciences & Engineering, 2019, 41(6):271.
[8] 袁梅, 商富凯, 董韶鹏. 基于经验小波变换的复合材料板声发射源定位[J]. 北京航空航天大学学报, 2018, 44(7): 1395-1401.
Yuan mei, Shang fukai, Dong shaopeng. Localization of Acoustic Emission Sources of Composite Panels Based on Empirical Wavelet Transform[J]. Journal of Beihang University 2018, 44(7): 1395-1401.
[9] Reusser R S, Holland S D, Chimenti D E, et al. Reflection and transmission of guided ultrasonic plate waves by vertical stiffeners[J]. The Journal of the Acoustical Society of America, 2014, 136(1): 170-182.
[10] Bian X, Yu Z, Shijiu J, et al. A new method of using sensor arrays for gas leakage location based on correlation of the time-space domain of continuous ultrasound. Sensors, 2015, 15(4): 8266-8283.
[11] 边旭，张宇，王佳强，等. 基于声传感器阵列的连续泄漏定位方法研究[J]. 振动与冲击, 2017, 36(6): 134-139.
Bian xu, Zhang yu, etc. Research on continuous leak location method based on acoustic sensor array[J]. Journal of Vibration and Shock, 2017, 36(6): 134-139.
[12] Tian S G, Ru D Q, Feng Z Y, et al. Investigation of artificial neural network pattern recognition of acoustic emission signals for pressure vessels[J]. Nonde Structive Testing, 2001(1): 37-58.
[13] Ahn B, Kim J, Choi B. Artificial intelligence-based machine learning considering flow and temperature of the pipeline for leak early detection using acoustic emission[J]. Engineering Fracture Mechanics, 2019, 210: 381-392.
[14] Wu Q, Lee C M. A modified leakage localization method using multilayer perceptron neural networks in a pressurized gas pipe[J]. Applied Sciences, 2019, 9(9): 1954.
[15] Niu H, Reeves E, Gerstoft P. Source localization in an ocean waveguide using supervised machine learning[J]. The Journal of the Acoustical Society of America, 2017, 142(3): 1176-1188.
[16] 牛海强，李整林，王海斌，等. 水声被动定位中的机器学习方法研究进展综述[J]. 信号处理, 2019, 35(9): 1450-1459.
Niu H, Li Zhenglin, etc. A review of the research progress of machine learning methods in underwater acoustic passive localization[J]. Signal processing, 2019, 35(9): 1450-1459.
[17] Rose Joseph L. Ultrasonic Waves in Solid Media [M]. Cambridge: Cambridge University Press, 2004.