Passive Sensing of Sensorized Composite Panels: Support Vector Machine

Abstract:

Strain readings recorded by surface mounted piezoelectric sensors due to impact events on composite panel are used to detect and characterize the impact. Sensor signals on a composite stiffened panels have been simulated by a valid numerical model. Applicability of least square support vector machines (LSSVM) on creating a meta-model to detect and characterize impact event has been investigated. In particular, the main advantage of LSSVM over other meta-modeling technique was found to be the smaller number of training data that is required. Experimental results on a composite panel has been used to validate the findings.

• Periodical: Key Engineering Materials (Volume 713)
• Pages: 199-202
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.713.199
• Online since: September 2016
• Authors: Nan Yue*, Zahra Sharif Khodaei, M.H. Aliabadi
• Keywords: Impact Characterization, Impact Detection, Machine Learning, Passive Sensing, Piezoelectric Sensor

References

[1] Ghajari, M., Z. Sharif-Khodaei, and M. Aliabadi. Impact detection using artificial neural networks. in Key Engineering Materials. 2012. Trans Tech Publ.

DOI: 10.4028/www.scientific.net/kem.488-489.767

[2] Sharif-Khodaei, Z., M. Ghajari, and M. Aliabadi, Determination of impact location on composite stiffened panels. Smart Materials and Structures, 2012. 21(10): p.105026.

DOI: 10.1088/0964-1726/21/10/105026

[3] Sharif-Khodaei, Z. and M. Aliabadi, Assessment of delay-and-sum algorithms for damage detection in aluminium and composite plates. Smart materials and structures, 2014. 23(7): p.075007.

DOI: 10.1088/0964-1726/23/7/075007

[4] Sharif-Khodaei, Z., O. Bacarreza, and M. Aliabadi. Lamb-Wave Based Technique for Multi-Site Damage Detection. in Key Engineering Materials. 2014. Trans Tech Publ.

DOI: 10.4028/www.scientific.net/kem.577-578.133

[5] Michaels, J.E., Detection, localization and characterization of damage in plates with an in situ array of spatially distributed ultrasonic sensors. Smart Materials and Structures, 2008. 17(3): p.035035.

DOI: 10.1088/0964-1726/17/3/035035

[6] Sharif-Khodaei, Z., M. Ghajari, and M. Aliabadi, Impact Damage Detection in Composite Plates using a Self-diagnostic Electro-Mechanical Impedance-based Structural Health Monitoring System. Journal of Multiscale Modelling, 2016: p.1550013.

DOI: 10.1142/s1756973715500134

[7] Ghajari, M., et al., Identification of impact force for smart composite stiffened panels. Smart Materials and Structures, 2013. 22(8): p.085014.

DOI: 10.1088/0964-1726/22/8/085014

[8] Suykens, J.A., Support vector machines: a nonlinear modelling and control perspective. European Journal of Control, 2001. 7(2): pp.311-327.

DOI: 10.3166/ejc.7.311-327

[9] Vapnik, V., The natural of statistical theory, 1995, New York: Springer-Verlag.

[10] Xie, J. Kernel optimization of LS-SVM based on damage detection for smart structures. in Computer Science and Information Technology, 2009. ICCSIT 2009. 2nd IEEE International Conference on. 2009. IEEE.

DOI: 10.1109/iccsit.2009.5234791

[11] Xu, Q., Impact detection and location for a plate structure using least squares support vector machines. Structural Health Monitoring, 2013: p.1475921713495083.

DOI: 10.1177/1475921713495083

[12] Olsson, R., Mass criterion for wave controlled impact response of composite plates. Composites Part A: Applied Science and Manufacturing, 2000. 31(8): pp.879-887.

DOI: 10.1016/s1359-835x(00)00020-8