损伤模式识别是空间碎片撞击航天器在轨感知技术中的一个重要功能模块,是目前研究的重点和难点。采用超高速撞击声发射技术,以铝合金平板为研究对象,通过大量超高速撞击实验获取实验信号,结合虚拟波阵面的精确源定位技术、时频分析技术及小波分解技术,从超高速撞击声发射信号中提取并优选与损伤模式直接相关的时频参数,建立了基于贝叶斯正则化BP神经网络的损伤模式识别方法,识别了铝合金板面受撞击形成的成坑/穿孔两种主要损伤模式。
Abstract
As a significant function module of space debris hypervelocity impact On-board monitoring technology, damage identification is the emphasis and difficulty of the research. Based on hypervelocity impact acoustic emission on aluminum plate, obtain varieties of hypervelocity impact acoustic emission signals by experiments. Combine with the accurate source location method with virtual wave front, specific time-frequency analysis and wavelet decomposition, extract and optimize the relevant parameters of damage pattern from hypervelocity impact acoustic emission signals, modeling a Bayesian Regularization BP neural network for damage pattern recognition method, carry out the pit and hole damage patterns recognition in aluminum plate.
关键词
空间碎片 /
超高速撞击 /
声发射 /
神经网络 /
损伤模式识别
{{custom_keyword}} /
Key words
space debris /
hypervelocity impact /
acoustic emission /
neural network /
damage pattern recognition
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Upper Stage Explosion Places LEO Satellites at Risk[N]. Orbital Debris Quarterly News..2013, 17(1): 8
[2] Corsaro R, Giovane F, Tsou P, et al. PINDROP An Acoustic Particle Impact Detector[N]. Orbital Debris Quarterly News. 2004, 8(3): 3-4.
[3] Prosser W H, Madaras E I. Distributed Impact Detector System (DIDS) Health Monitoring System Evaluation[R]. Hampton: Langley Research Center, 2010.
[4] Spencer G, Schäfer F, Tanaka M, et al. Design and initial calibration of micrometeoroid space debris detector (MDD)[Z]. Darmstadt, Germany: 2005.
[5] 熊秋鹏. 基于神经网络技术的空间碎片损伤模式识别研究[D]. 哈尔滨: 哈尔滨工业大学硕士学位论文, 2012: 34-36.
Xiong Qiupeng. Research on Space Debris Damage Pattern Recognition Based on Neutral Network Technology[D]. Harbin: Harbin Institute of Technology, 2012: 34~36.
[6] Liu Wugang, Pang Baojun, Han Zengyao, Sun Fei. Acoustic emission detection and location for hypervelocity impacts based on wavelet transform[J]. Gaojishu Tongxin/Chinese High Technology Letters, 2009; (19): 181-187.
[7] Liu Z.D., Pang B J.A method based on acoustic emission for locating debris cloud impact[C]. 4th International Conference on Experimental Mechanics . 2009.
[8] 张凯, 庞宝君, 林敏. 碎片云撞击声发射信号能量特征小波包分析[J]. 振动与冲击. 2012, vol.31, (12): 125-128.
Zhang Kai, Pang Baojun, Lin Min. Wavelet packet analysis for acoustic emission signals caused by debris cloud impact[J]. Journal of Vibration and shock. 2012, vol.31, N(12): 125-128.
[9] Ditri, J.,Rose. J.L., and Chen, G. . Mode selection guidelines for defect detection optimization using Lamb waves[C]. Proceeding of the 18th Annual Review of progress in Quantitative NDE, 1991, vol.11, 2109-2115. NewYork.
[10] Mackay D. J. C., Bayesian interpolation. Neural Computation[J], 1992; 4:415—447.
[11] R. Orre, A. Lansner, A. Bate, M.Lindquist. Bayesian neural networks with confidence estimations applied to data mining[J]. Computational Statistics & Data Analysis, 2000, (34),:473-493.
[12] Man Leung Wong, Shing Yan Lee, K. Wong Sak Leung. Data mining of Bayesian networks using cooperative coevolution[J]. Decision Support Systems, 2004, (8):451-472.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}