Page 39 - 《应用声学》2021年第3期

第 40 卷 第 3 期 张洪等： 应用深度学习识别法兰螺栓连接状态 357

[5] 张健奎, 王宁, 卢萍, 等. 辨识振动环境中两点螺栓连接状态 [11] Jung B H, Kim Y W, Lee J R. Laser-based struc-
的 REE 声发射指标 [J]. 振动与冲击, 2013, 32(8): 179–182. tural training algorithm for acoustic emission localization
Zhang Jiankui, Wang Ning, Lu Ping, et al. AE REE and damage accumulation visualization in a bolt joint
index to identify connecting state of a 2-bolt connected structure[J]. Structural Health Monitoring, 2019, 18(5/6):
structure[J]. Journal of Vibration and Shock, 2013, 32(8): 1851–1861.
179–182. [12] Zhang Y, Sun X, Loh K J, et al. Autonomous bolt loos-
[6] 张陆佳, 林兰天, 陈春敏, 等. 基于主成分分析的纤维拉伸断 ening detection using deep learning[J]. Structural Health
裂声发射信号识别 [J]. 纺织学报, 2018, 39(1): 19–24. Monitoring, 2020, 19(1): 105–122.
Zhang Lujia, Lin Lantian, Chen Chunmin, et al. Iden- [13] Zhao X, Zhang Y, Wang N. Bolt loosening angle detection
tification of fiber tensile fracture acoustic emission signal technology using deep learning[J]. Structural Control and
based on principal component analysis[J]. Journal of Tex- Health Monitoring, 2019, 26(1): e2292.1–e2292.14.
tile Research, 2018, 39(1): 19–24. [14] 黄金, 吴庆良, 陈钒. 基于 CEEMDAN-WPT 联合去噪的灾
[7] Vongserewattana N, Suwansin W, Phasukkit P, et al. 后求救信号能量分布特征研究 [J]. 南京理工大学学报 (自然科
Validation of acoustic emission railway track crack anal- 学版), 2020, 44(2): 194–201.
ysis using MFCC[C]// 2019 16th International Confer- Huang Jin, Wu Qingliang, Chen Fan. Study on energy
ence on Electrical Engineering/Electronics, Computer, distribution character about post-disaster rescue signal
Telecommunicationsand Information Technology (ECTI- based on CEEMDAN-WPT denoising[J]. Journal of Nan-
CON), IEEE, 2019: 633–636. jing University of Science and Technology, 2020, 44(2):
[8] 司莉, 毕贵红, 魏永刚, 等. 基于 RQA 与 SVM 的声发射信号 194–201.
检测识别方法 [J]. 振动与冲击, 2016, 35(2): 97–103, 123. [15] 陆彦希, 曹乐. 基于改进 CEEMDAN 和 TEO 的轴承故障特
Si Li, Bi Guihong, Wei Yonggang, et al. Detection and 征提取方法 [J]. 噪声与振动控制, 2020, 40(2): 109–114.
identification of acoustic emission signals based on re- Lu Yanxi, Cao Le. Bearing fault feature extraction
currence quantification analysis and support vector ma- method based on improved CEEMDAN and TEO[J].
chines[J]. Journal of Vibration and Shock, 2016, 35(2): Noise and Vibration Control, 2020, 40(2): 109–114.
97–103, 123. [16] Ai O C, Hariharan M, Yaacob S, et al. Classifica-
[9] Shin H C, Roth H R, Gao M, et al. Deep convo- tion of speech dysfluencies with MFCC and LPCC fea-
lutional neural networks for computer-aided detection: tures[J]. Expert Systems with Applications, 2012, 39(2):
CNN architectures, dataset characteristics and transfer 2157–2165.
learning[J]. IEEE Transactions on Medical Imaging, 2016, [17] Simonyan K, Zisserman A. Very deep convolutional net-
35(5): 1285–1298. works for large-scale image recognition[J]. arXiv preprint,
[10] Babu G S, Zhao P, Li X L. Deep convolutional neural net- arXiv: 1409.1556, 2014.
work based regression approach for estimation of remain- [18] San-Segundo R, Gil-Martín M, D’Haro-Enríquez L F, et
ing useful life[C]. International Conference on Database al. Classification of epileptic EEG recordings using signal
Systems for Advanced Applications. Springer, Cham, transforms and convolutional neural networks[J]. Com-
2016: 214–228. puters in Biology and Medicine, 2019, 109: 148–158.