Smart Structures and Systems

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Data abnormal detection using bidirectional long-short neural network combined with artificial experience

  • Yang, Kang (School of Civil Engineering, Key Laboratory of C&PC Structures of the Ministry of Education, Southeast University) ;
  • Jiang, Huachen (School of Civil Engineering, Key Laboratory of C&PC Structures of the Ministry of Education, Southeast University) ;
  • Ding, Youliang (School of Civil Engineering, Key Laboratory of C&PC Structures of the Ministry of Education, Southeast University) ;
  • Wang, Manya (School of Civil Engineering, Key Laboratory of C&PC Structures of the Ministry of Education, Southeast University) ;
  • Wan, Chunfeng (School of Civil Engineering, Key Laboratory of C&PC Structures of the Ministry of Education, Southeast University)
  • Received : 2021.04.14
  • Accepted : 2021.07.12
  • Published : 2022.01.25
⟨ Previous Next ⟩

Abstract

Data anomalies seriously threaten the reliability of the bridge structural health monitoring system and may trigger system misjudgment. To overcome the above problem, an efficient and accurate data anomaly detection method is desiderated. Traditional anomaly detection methods extract various abnormal features as the key indicators to identify data anomalies. Then set thresholds artificially for various features to identify specific anomalies, which is the artificial experience method. However, limited by the poor generalization ability among sensors, this method often leads to high labor costs. Another approach to anomaly detection is a data-driven approach based on machine learning methods. Among these, the bidirectional long-short memory neural network (BiLSTM), as an effective classification method, excels at finding complex relationships in multivariate time series data. However, training unprocessed original signals often leads to low computation efficiency and poor convergence, for lacking appropriate feature selection. Therefore, this article combines the advantages of the two methods by proposing a deep learning method with manual experience statistical features fed into it. Experimental comparative studies illustrate that the BiLSTM model with appropriate feature input has an accuracy rate of over 87-94%. Meanwhile, this paper provides basic principles of data cleaning and discusses the typical features of various anomalies. Furthermore, the optimization strategies of the feature space selection based on artificial experience are also highlighted.

Keywords

Acknowledgement

The authors would like to thank the organizations of the International Project Competition for SHM (IPC-SHM 2020) ANCRiSST, Harbin Institute of Technology (China), and University of Illinois at Urbana-Champaign (USA) for their generously providing the invaluable data from actual structures. The authors also would like to thank the chairs of IPC-SHM 2020 Prof. Hui Li, and Prof. Billie F. Spencer Jr. for their leadership on the competition. Also, the authors sincerely acknowledge financial support from the National Natural Science Foundation of China (Grants. 51978154), the Fund for Distinguished Young Scientists of Jiangsu Province (Grant. BK20190013), Key Research and Development Program of Nanjing Jiangbei New Area (Grant. ZDYF20200118) and Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX21_0113).

References

  1. Bao, Y. and Li, H. (2020), "Machine learning paradigm for structural health monitoring", Struct. Health Monitor., 20(4), 1353-1372. https://doi.org/10.1177/1475921720972416
  2. Bao, Y., Chen Z., Wei, S., Xu, Y., Tang, Z. and Li, H. (2019a), "The state of the art of data science and engineering in structural health monitoring", Engineering, 5(2), 234-242. https://doi.org/10.1016/j.eng.2018.11.027
  3. Bao, Y., Tang, Z., Li, H. and Zhang, Y. (2019b), "Computer vision and deep learning-based data anomaly detection method for structural health monitoring", Struct. Health Monitor., 18(2), 401-421. https://doi.org/10.1177/1475921718757405
  4. Chen, Z., Bao, Y., Li, H. and Spencer, B. (2018a), "A novel distribution regression approach for data loss compensation in structural health monitoring", Struct. Health Monitor., 17(6), 1473-1490. https://doi.org/10.1177/1475921717745719
  5. Chen, Z., Li, H. and Bao, Y. (2018b), "Analyzing and modeling inter-sensor relationships for strain monitoring data and missing data imputation: a copula and functional data-analytic approach", Struct. Health Monitor., 18(4), 1168-1188. https://doi.org/10.1177/1475921718788703
  6. Chen, Z., Bao, Y., Li, H. and Spencer, B. (2019), "LQD-RKHS-based distribution-to-distribution regression methodology for restoring the probability distributions of missing SHM data", Mech. Syst. Signal Process., 121, 655-674. https://doi.org/10.1016/j.ymssp.2018.11.052
  7. Hochreiter, S. and Schmidhuber, J. (1997), "Long short-term memory", Neural Comput., 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  8. Huang, Y., Beck, J.L., Wu, S. and Li, H. (2016), "Bayesian compressive sensing for approximately sparse signals and application to structural health monitoring signals for data loss recovery", Probabil. Eng. Mech., 46, 62-79. https://doi.org/10.1016/j.probengmech.2016.08.001
  9. Li, L., Liu, G., Zhang, L. and Li, Q. (2019), "Sensor fault detection with generalized likelihood ratio and correlation coefficient for bridge SHM", J. Sound Vib., 442, 445-458. https://doi.org/10.1016/j.jsv.2018.10.062
  10. Liu, G., Li, L., Zhang, L., Li, Q. and Law, S.S. (2020), "Sensor faults classification for SHM systems using deep learning-based method with Tsfresh features", Smart Mater. Struct., 29(7), 075005. https://iopscience.iop.org/article/10.1088/1361-665X/ab85a6
  11. Liu, H., Shah, S. and Jiang, W. (2004), "On-line outlier detection and data cleaning", Comput. Chem. Eng., 28(9), 1635-1647. https://doi.org/10.1016/j.compchemeng.2004.01.009
  12. Kullaa, J. (2013), "Detection, identification, and quantification of sensor fault in a sensor network", Mech. Syst. Signal Process., 40(1), 208-221. https://doi.org/10.1016/j.ymssp.2013.05.007
  13. Ni, F., Zhang, J. and Noori, M.N. (2019), "Deep learning for data anomaly detection and data compression of a long-span suspension bridge", Comput.-Aided Civil Inf., 35(7), 685-700. https://doi.org/10.1111/mice.12528
  14. Schuster, M. and Paliwal, K.K. (1997), "Bidirectional recurrent neural networks", IEEE T Signal Process., 45(11), 2673-2681. https://ieeexplore.ieee.org/document/650093 https://doi.org/10.1109/78.650093
  15. Spencer Jr., B.F., Hoskere, V. and Narazaki, Y. (2019), "Advances in computer vision-based civil infrastructure inspection and monitoring", Engineering, 5(2), 199-222. https://doi.org/10.1016/j.eng.2018.11.030
  16. Tang, Z., Chen, Z., Bao, Y. and Li, H. (2019), "Convolutional neural network-based data anomaly detection method using multiple information for structural health monitoring", Struct. Control Health., 26(1), e2296. https://onlinelibrary.wiley.com/doi/abs/10.1002/stc.2296
  17. Xia, Y. and Ni, Y. (2018), "A wavelet-based despiking algorithm for large data of structural health monitoring", Int. J. Distrib. Sens. N, 14(12),1550147718819095. https://doi.org/10.1177/1550147718819095
  18. Yang, Y. and Nagarajaiah, S. (2016), "Harnessing data structure for recovery of randomly missing structural vibration responses time history: sparse representation versus low-rank structure", Mech. Syst. Signal Process., 74, 165-182. https://doi.org/10.1016/j.ymssp.2015.11.009
  19. Yi, T., Huang, H. and Li, H. (2017), "Development of sensor validation methodologies for structural health monitoring: a comprehensive review", Measurement, 109, 200-214. https://doi.org/10.1016/j.measurement.2017.05.064
  20. Yu, M., Wang, D. and Luo, M. (2013), "Model-based prognosis for hybrid systems with mode-dependent degradation behaviors", IEEE Trans. Ind. Electron., 61(1), 546-554. https://ieeexplore.ieee.org/document/6425465 https://doi.org/10.1109/TIE.2013.2244538
  21. Yuen, K.-V. and Ortiz, G.A. (2017), "Outlier detection and robust regression for correlated data", Comput. Method Appl. Mech. Eng., 313, 632-646. https://doi.org/10.1016/j.cma.2016.10.004