Dimensionality Reduction in Modal Analysis for Structural Health Monitoring

Autonomous structural health monitoring (SHM) of many structures and bridges became a topic of paramount importance for maintenance purposes and safety reasons. This paper proposes a set of machine learning (ML) tools to perform automatic feature selection and detection of anomalies in a bridge from vibrational data and compare different feature extraction schemes to increase the accuracy and reduce the amount of data collected. As a case study, the Z-24 bridge is considered because of the extensive database of accelerometric data in both standard and damaged conditions. The proposed framework starts from the first four fundamental frequencies extracted through operational modal analysis (OMA) and clustering, followed by time-domain filtering (tracking). The fundamental frequencies extracted are then fed to a dimensionality reduction block implemented through two different approaches: feature selection (intelligent multiplexer) that tries to estimate the most reliable frequencies based on the evaluation of some statistical features (i.e., entropy, variance, kurtosis), and feature extraction (auto-associative neural network (ANN)) that combine the fundamental frequencies to extract new damage sensitive features in a low dimensional feature space. Finally, one-class classification (OCC) algorithms perform anomaly detection, trained with standard condition points, and tested with normal and anomaly ones. In particular, principal component analysis (PCA), kernel principal component analysis (KPCA), and autoassociative neural network (ANN) are presented and their performance are compared. It is also shown that, by evaluating the correct features, the anomaly can be detected with accuracy and an F1 score greater than 95%.





References:
[1] R. Ferrari, D. Froio, E. Chatzi, F. Pioldi, and E. Rizzi, “Experimental and
numerical investigations for the structural characterization of a historic
RC arch bridge,” in Proc. Int. Conf. on Comp. Methods in Structural
Dynamics and Earthquake Eng., vol. 1, Athens, Greece, May 2015, pp.
2337–2353.
[2] A. Benedetti, M. Tarozzi, G. Pignagnoli, and C. Martinelli, “Dynamic
investigation and short-monitoring of an historic multi-span masonry
arch bridge,” in Proc. Int. Conf. on Arch Bridges, vol. 11, Porto, Portugal,
Oct. 2019, pp. 831–839.
[3] A. Benedetti, G. Pignagnoli, and M. Tarozzi, “Damage identification of
cracked reinforced concrete beams through frequency shift,” Materials
and Structures, vol. 51, pp. 1–15, Oct. 2018.
[4] A. Benedetti, C. Colla, G. Pignagnoli, and M. Tarozzi, “Static and
dynamic investigation of the taro masonry bridge in parma, italy,” in
Proc. Int. Conf. on Struct. Analysis of Hist. Const., vol. 18, Cusco,
Per`u, Sep. 2019, pp. 2264–2272.
[5] G. D. Roeck, “The state-of-the-art of damage detection by vibration
monitoring: the SIMCES experience,” J. of Structural Control, vol. 10,
no. 2, pp. 127–134, May 2003.
[6] K. Worden, C. Farrar, J. Haywood, and M. Todd, “A review of nonlinear
dynamics applications to structural health monitoring,” Structural
Control and Health Monitoring, vol. 15, no. 4, pp. 540–567, Jul. 2008.
[7] G. Fabbrocino and C. Rainieri, Operational modal analysis of civil
engineering structures. New York: Springer-verlag, May 2014.
[8] S. Taylor, K. Farinholt, E. Flynn, E. Figueiredo, D. L. Mascarenas, E. A.
Moro, G. Park, M. Todd, and C. Farrar, “A mobile-agent-based wireless
sensing network for structural monitoring applications,” Measurement
Science and Technology, vol. 20, no. 4, pp. 1–14, Jan. 2009.
[9] C. M. Bishop, Pattern Recognition and Machine Learning. Springer
Verlag, Aug. 2006.
[10] J. Watt, R. Borhani, and A. K. Katsaggelos, Machine Learning Refined.
Cambridge University Press, 2016.
[11] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press,
2016.
[12] E. Favarelli and A. Giorgetti, “Machine learning for automatic
processing of modal analysis in damage detection of bridges,” IEEE
Trans. on Instrumentation and Measurement, 2020.
[13] L. Pucci, E. Testi, E. Favarelli, and A. Giorgetti, “Human activities
classification using biaxial seismic sensors,” IEEE Sensors Letters,
vol. 4, no. 10, pp. 1 – 4, Oct. 2020.
[14] Z-24 bridge data. [Online]. Available: https://bwk.kuleuven.be/bwm/z24
[15] E. P. Carden and J. M. W. Brownjohn, “Fuzzy clustering of
stability diagrams for vibration-based structural health monitoring,”
Computer-Aided Civil and Infrastructure Eng., vol. 23, no. 5, pp.
360–372, May 2008.
[16] C. Wu, H. Liu, X. Qin, and J. Wang, “Stabilization diagrams to
distinguish physical modes and spurious modes for structural parameter
identification,” Journal of Vibroeng., vol. 19, no. 4, pp. 2777–2794, Jun.
2017. [17] D. Brigante, C. Rainieri, and G. Fabbrocino, “The role of the modal
assurance criterion in the interpretation and validation of models for
seismic analysis of architectural complexes,” in Proc. Int. Conf. on
Structural Dynamics (Eurodin), vol. 199, Rome, Italy, Sep. 2017, pp.
3404–3409.
[18] A. Cabboi, F. Magalha˜es, C. Gentile, and A´ . Cunha, “Automated
modal identification and tracking: Application to an iron arch bridge,”
Structural Control and Health Monitoring, vol. 24, no. 1, p. e1854, Feb.
2017.
[19] M. Pastor, M. Binda, and T. Harˇcarik, “Modal assurance criterion,”
Procedia Engineering, vol. 48, pp. 543–548, 2012.
[20] E. Reynders and G. D. Roeck, “Continuous vibration monitoring and
progressive damage testing on the z 24 bridge,” Encyclopedia of
structural health monitoring, vol. 10, pp. 127–134, May 2009.
[21] E. Reynders, J. Houbrechts, and G. D. Roeck”, “Fully automated
(operational) modal analysis,” Mechanical Systems and Signal
Processing, vol. 29, pp. 228–250, May 2012.
[22] A. Santos, M. Silva, C. Sales, J. Costa, and E. Figueiredo, “Applicability
of linear and nonlinear principal component analysis for damage
detection,” in Proc. IEEE Int. Instr. and Meas. Tech. Conf. (I2MTC),
Pisa, Italy, May 2015, pp. 869–874.
[23] E. Favarelli, E. Testi, L. Pucci, and A. Giorgetti, “Anomaly detection
using wifi signal of opportunity,” in Proc. IEEE Int. Conf. on Signal
Proc. and Comm. Sys. (ICSPCS), Surfers Paradise, Gold Coast,
Australia, Dec. 2019, pp. 1–7.
[24] E. Favarelli, E. Testi, and A. Giorgetti, “One class classifier neural
network for anomaly detection in low dimensional feature spaces,” in
Proc. IEEE Int. Conf. on Signal Proc. and Comm. Sys. (ICSPCS), Surfers
Paradise, Gold Coast, Australia, Dec. 2019, pp. 1–7.
[25] P. Perera and V. M. Patel, “Learning deep features for one-class
classification,” IEEE Trans. Image Process, vol. 28, no. 11, pp.
5450–5463, Nov. 2019.
[26] R. Chalapathy, A. K. Menon, and S. Chawla, “Anomaly detection using
one-class neural networks,” CoRR, vol. abs/1802.06360, Aug. 2018.
[27] H. Abdi and L. J. Williams, “Principal component analysis,” Wiley
Interd. Reviews: Comp. Stat., vol. 2, no. 4, pp. 433–459, 2010.
[28] B. Sch¨olkopf, A. Smola, and K.-R. M¨uller, “Kernel principal component
analysis,” in Proc. Int. conf. on artificial neural networks, vol. 1327,
no. 6. Lausanne, Switzerland: Springer, Oct. 1997, pp. 583–588.
[29] B. Sch¨olkopf, A. Smola, E. Smola, and K.-R. M¨uller, “Nonlinear
component analysis as a kernel eigenvalue problem,” Neural Comp.,
vol. 10, pp. 1299–1319, Jul. 1998.
[30] M. Silva, A. Santos, E. Figueiredo, R. Santos, C. Sales, and J. Costa, “A
novel unsupervised approach based on a genetic algorithm for structural
damage detection in bridges,” Eng. Applications of Artificial Intelligence,
vol. 52, pp. 168–180, Jun. 2016.
[31] M. Silva, A. Santos, R. Santos, E. Figueiredo, C. Sales, and J. C. Costa,
“Agglomerative concentric hypersphere clustering applied to structural
damage detection,” Mechanical Systems and Signal Processing, vol. 92,
pp. 196–212, Feb. 2017.