Infrared and Laser Engineering, Volume. 53, Issue 11, 20240294(2024)
Cluster-based recognition method for Φ-OTDR system's vibration signals
Figures & Tables(25)
Fig. 4. (a) Original data graph of manual percussion; (b) Original data graph of machine excavation; (c) Cropped data graph of manual percussion; (d) Cropped data graph of machine excavation; (e) Envelope value graph of manual percussion; (f) Envelope value graph of machine excavation
Fig. 6. (a) Three-dimensional projection of the original data after dimensionality reduction; (b) Three-dimensional plot of the effect of agglomerative clustering
Fig. 7. Line charts of normalized time-domain characteristics of wind noise. (a) Line chart with 15 characteristic values; (b)
Fig. 8. Line charts of normalized time-domain characteristics for manual knocking. (a) Line chart with 15 characteristic values; (b)
Fig. 9. Line charts of normalized time-domain characteristics for machine excavation. (a) Line chart with 15 characteristic values; (b) Feature values of
Fig. 11. Elbow diagrams for agglomerative clustering of cropped normalized data
Fig. 12. Under
Fig. 13. Evaluation coefficients for agglomerative clustering under
Fig. 14. Confusion matrix for agglomerative clustering under
Fig. 15. Under the condition of
Fig. 16. Confusion matrix for agglomerative clustering under the condition of
Fig. 17. (a) Plot of rainfall data; (b) Plot of windblown data; (c) Plot of direct tapping data; (d) Plot of indirect tapping data
Fig. 18. Cluster evaluation coefficients of this study and previous studies
Table 1. Characterization of the significance of the seven time-domain features
View tableView in ArticleTable 1. Characterization of the significance of the seven time-domain features
No. Time domain feature Characterization meaning 1 Maximum value Represents the peak amplitude of the optical fiber vibration signal 2 Mean value Represents the average amplitude of the optical fiber vibration signal 3 Variance Represents the power of the optical fiber vibration signal 4 Skewness Represents the symmetry of the optical fiber vibration signal 5 Waveform factor Represents the shape of the optical fiber vibration signal 6 Crest factor Represents the degree of peak protrusion of the optical fiber vibration signal 7 Impulse factor Represents the impulse characteristics of the optical fiber vibration signal
Table 2. Cluster assessment indicators
View tableView in ArticleTable 2. Cluster assessment indicators
Indicators Categories Description Homogeneity External evaluation Measures the similarity of data points within a cluster Completeness External evaluation Measures the allocation degree of data points with the same characteristics V-measure External evaluation Comprehensively considers homogeneity and completeness Adjustedrand index External evaluation Measures the consistency between clustering results and true labels in case of randomness Adjusted mutual information External evaluation Measures the correlation between clustering results and true labels Silhouette coefficient Internal evaluation Measures the suitability of the distribution of data points within each cluster
Table 3. Parameters of devices in experiment
View tableView in ArticleTable 3. Parameters of devices in experiment
Parameter Value Center wavelength of laser/nm 1550.92 Laser linewidth/kHz 5 Modulator modulation pulse width/ns 100 Modulator pulse interval/ms 0.02 Acquisition card sampling rate/106 samples·s−1 90 System spatial resolution/m 10 Experimental optical cable length/m 2 000
Table 4. Time domain features corresponding to symbols
View tableView in ArticleTable 4. Time domain features corresponding to symbols
No. Symbol Time domain characteristics 1 Bmax Maximum value 2 Bmin Minimum value 3 Bmeans Average value 4 Bvar Variance 5 Bav Standard deviation 6 Bff Peak-to-peak value 7 Bav Rectified average value 8 Bku Kurtosis 9 Bsk Skewness 10 Brm Root mean square of the mean 11 Bs Waveform factor 12 Bc Crest factor 13 Bi Impulse factor 14 Bl Margin factor 15 Ben Signal entropy
Table 5. Number of input samples and number of clusters for various events
View tableView in ArticleTable 5. Number of input samples and number of clusters for various events
Event Number of input samples Number of cluster samples Total accuracy Wind noise 504 504 88.68% Manual knock 548 543 Machine excavation 591 596
Table 6. Number of input samples, number of samples correctly clustered and accuracy under different
K conditionsView tableView in ArticleTable 6. Number of input samples, number of samples correctly clustered and accuracy under different
K conditionsK setting value Input sample size Number of samples correctly clustered Accuracy 2 1643 1643 100% 3 1643 1453 88.68% 4 1643 1250 76.26%
Table 7. Number of samples for three correct clusters under different
K conditionsView tableView in ArticleTable 7. Number of samples for three correct clusters under different
K conditionsK setting value Number of correctly clustered samples for wind noise Number of correctly clustered samples for manual knock Number of correctly clustered samples for excavation 2 504 1139 3 504 477 434 4 504 438 295
Tools
Get Citation
Copy Citation Text
Nianchao LIU, Qin LI, Xiaoting ZHAO, Sheng LIANG. Cluster-based recognition method for Φ-OTDR system's vibration signals[J]. Infrared and Laser Engineering, 2024, 53(11): 20240294
Paper Information
Category: Optical communication and sensing
Received: Jul. 21, 2024
Accepted: --
Published Online: Dec. 13, 2024
The Author Email: LI Qin (liqin_buaa@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20