Restricted access
Research articles
Research articlesMachine learning at the interface of structural health monitoring and non-destructive evaluation
Published:14 September 2020https://doi.org/10.1098/rsta.2019.0581
Abstract
While both non-destructive evaluation (NDE) and structural health monitoring (SHM) share the objective of damage detection and identification in structures, they are distinct in many respects. This paper will discuss the differences and commonalities and consider ultrasonic/guided-wave inspection as a technology at the interface of the two methodologies. It will discuss how data-based/machine learning analysis provides a powerful approach to ultrasonic NDE/SHM in terms of the available algorithms, and more generally, how different techniques can accommodate the very substantial quantities of data that are provided by modern monitoring campaigns. Several machine learning methods will be illustrated using case studies of composite structure monitoring and will consider the challenges of high-dimensional feature data available from sensing technologies like autonomous robotic ultrasonic inspection.
This article is part of the theme issue ‘Advanced electromagnetic non-destructive evaluation and smart monitoring’.
Footnotes
References
Reference
- 1.
Farrar C, Worden K . 2012Structural health monitoring: a machine learning perspective. Chichester, UK: John Wiley and Sons. Crossref, Google Scholar - 2.
Worden K, Dulieu-Barton J . 2004Intelligent damage identification in systems and structures. Int. J. Struct. Health Monit. 3, 85–98. (doi:10.1177/1475921704041866) Crossref, Google Scholar - 3.
Rytter A . 1993Vibrational based inspection of civil engineering structures. PhD. thesis, Aalborg University, Denmark. Google Scholar - 4.
Cawley P . 2000Long-range inspection of structures using low frequency ultrasound. In Proc. of 2nd Int. Workshop on Damage Assessment using Advanced Signal Processing Procedures – DAMAS ’97, Sheffield, UK, pp. 1–17. Google Scholar - 5.
Kessler S, Spearing S, Soutis C . 2002Damage detection in composite materials using Lamb wave methods. Smart Mater. Struct. 11, 269–278. (doi:10.1088/0964-1726/11/2/310) Crossref, ISI, Google Scholar - 6.
Croxford A, Wilcox P, Drinkwater B, Konstantinidis G . 2007Strategies for guided-wave structural health monitoring. Proc. R. Soc. A: Math., Phys. Eng. Sci. 463, 2961–2981. (doi:10.1098/rspa.2007.0048) Link, ISI, Google Scholar - 7.
McNab A, Campbell J . 1987Ultrasonic phased arrays for nondestructive testing. NDT Int. 20, 333–337. (doi:10.1016/0308-9126(87)90290-2) ISI, Google Scholar - 8.
Aranguren G, Monjel P, Cokonaj V, Barrera E, Ruiz M . 2013Ultrasonic wave-based structural health monitoring embedded instrument. Rev. Sci. Instrum. 84, 125106. (doi:10.1063/1.4834175) Crossref, PubMed, ISI, Google Scholar - 9.
Li M, Hayward G . 2012Ultrasound nondestructive evaluation (NDE) imaging with transducer arrays and adaptive processing. Sensors 12, 42–54. (doi:10.3390/s120100042) Crossref, ISI, Google Scholar - 10.
Chiao R, Thomas L . 1994Analytical evaluation of sampled aperture ultrasonic imaging techniques for NDE. IEEE Trans. Ultrason., Ferroelectr. Freq. Control 41, 484–493. (doi:10.1109/58.294109) Crossref, ISI, Google Scholar - 11.
Holmes C, Drinkwater B, Wilcox P . 2005Post-processing of the full matrix of ultrasonic transmit-receive array data for non-destructive evaluation. NDT & E Int. 38, 701–711. (doi:10.1016/j.ndteint.2005.04.002) Crossref, ISI, Google Scholar - 12.
Flynn E, Chong S, Jarmer G, Lee JR . 2013Structural imaging through local wavenumber estimation of guided waves. NDT & E Int. 59, 1–10. (doi:10.1016/j.ndteint.2013.04.003) Crossref, ISI, Google Scholar - 13.
Jeon J, Gang S, Park G, Flynn E, Kang T, Han S . 2017Damage detection on composite structures with standing wave excitation and wavenumber analysis. Adv. Composite Mater. 26, 53–65. (doi:10.1080/09243046.2017.1313577) Crossref, ISI, Google Scholar - 14.
Murphy K . 2012Machine learning: a probabilistic perspective. Cambridge, MA: MIT press. Google Scholar - 15.
Manson G, Worden K, Allman D . 2003Experimental validation of a structural health monitoring methodology: Part III. Damage location on an aircraft wing. J. Sound Vib. 259, 365–385. (doi:10.1006/jsvi.2002.5169) Crossref, ISI, Google Scholar - 16.
Janssens O, Van de Walle R, Loccufier M, Van Hoecke S . 2018Deep learning for infrared thermal image based machine health monitoring. IEEE/ASME Trans. Mechatron. 23, 151–159. (doi:10.1109/TMECH.2017.2722479) Crossref, ISI, Google Scholar - 17.
Zhao R, Yan R, Chen Z, Mao K, Wang P, Gao RX . 2019Deep learning and its applications to machine health monitoring. Mech. Syst. Signal Process. 115, 213–237. (doi:10.1016/j.ymssp.2018.05.050) Crossref, ISI, Google Scholar - 18.
Bornn L, Farrar CR, Park G, Farinholt K . 2009Structural health monitoring with autoregressive support vector machines. J. Vib. Acoust. 131, 021004. (doi:10.1115/1.3025827) Crossref, ISI, Google Scholar - 19.
Fuentes R, Cross EJ, Halfpenny A, Worden K, Barthorpe RJ . 2014Aircraft parametric structural load monitoring using gaussian process regression. In In the Proc. 7th European Workshop on Structural Health Monitoring. Google Scholar - 20.
Holmes G, Sartor P, Reed S, Southern P, Worden K, Cross E . 2016Prediction of landing gear loads using machine learning techniques. Struct. Health Monit. 15, 568–582. (doi:10.1177/1475921716651809) Crossref, Google Scholar - 21.
Rogers TJ, Worden K, Fuentes R, Dervilis N, Tygesen UT, Cross EJ . 2019A Bayesian non-parametric clustering approach for semi-supervised structural health monitoring. Mech. Syst. Signal Process. 119, 100–119. (doi:10.1016/j.ymssp.2018.09.013) Crossref, ISI, Google Scholar - 22.
Bull LA, Rogers TJ, Wickramarachchi C, Cross EJ, Worden K, Dervilis N . 2019Probabilistic active learning: an online framework for structural health monitoring. Mech. Syst. Signal Process. 134, 106294. (doi:10.1016/j.ymssp.2019.106294) Crossref, ISI, Google Scholar - 23.
Fuentes R, Gardner P, Mineo C, Rogers TJ, Pierce SG, Worden K, Dervilis N, Cross EJ . 2020Autonomous ultrasonic inspection using Bayesian optimisation and robust outlier analysis. Mech. Syst. Signal Process. 145, 106897. (doi:10.1016/j.ymssp.2020.106897) Crossref, ISI, Google Scholar - 24.
Fuentes R, Mineo C, Pierce SG, Worden K, Cross EJ . 2019A probabilistic compressive sensing framework with applications to ultrasound signal processing. Mech. Syst. Signal Process. 117, 383–402. (doi:10.1016/j.ymssp.2018.07.036) Crossref, ISI, Google Scholar - 25.
Kurz JH, Grosse CU, Reinhardt H . 2005Strategies for reliable automatic onset time picking of acoustic emissions and of ultrasound signals in concrete. Ultrasonics 43, 538–546. (doi:10.1016/j.ultras.2004.12.005) Crossref, PubMed, ISI, Google Scholar - 26.
Legendre S, Goyette J, Massicotte D . 2001Ultrasonic nde of composite material structures using wavelet coefficients. NDT & E Int. 34, 31–37. (doi:10.1016/S0963-8695(00)00029-3) Crossref, ISI, Google Scholar - 27.
Cardoso G, Saniie J . 2015Data compression and noise suppression of ultrasonic NDE signals using wavelets. In IEEE Symp. on Ultrasonics, volume 16, pp. 250–253. Google Scholar - 28.
Tipping ME . 2000The relevance vector machine. In Advances in Neural Information Processing Systems, pp. 652–658. Cambridge, MA: MIT Press. Google Scholar - 29.
Johnson WB, Lindenstrauss J . 1984Extensions of Lipschitz mapping into a hilbert space. Contemp. Math. 26, 189–206. (doi:10.1090/conm/026/7374000) Crossref, Google Scholar - 30.
- 31.
Bogue R . 2010The role of robotics in non-destructive testing. Ind. Robot 37, 421–426. (doi:10.1108/01439911011063236) Crossref, ISI, Google Scholar - 32.
Mineo C, MacLeod C, Morozov M, Pierce G, Summan R, Rodden T, Kahani D, Powell J, McCubbin P, McCubbin C, Munro G, Paton S, Watson D . 2017Flexible integration of robotics ultrasonics and metrology for the inspection of aerospace components. In AIP Conf. Proc., vol. 1806, p. 020026. Crossref, Google Scholar - 33.
Rousseeuw P, Van Driessen K . 1999A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223. (doi:10.1080/00401706.1999.10485670) Crossref, ISI, Google Scholar - 34.
Rasmussen C, Williams C . 2006Gaussian processes for machine learning. Cambridge, MA: MIT Press. Google Scholar - 35.
Jones D, Schonlau M, Welch W . 1998Efficient global optimization of expensive black-box functions. J. Global Optim. 13, 455–492. (doi:10.1023/A:1008306431147) Crossref, ISI, Google Scholar - 36.
Gardner P, Liu X, Worden K . 2019On the application of domain adaptation in structural health monitoring. Mech. Syst. Signal Process. 138, 106550. (doi:10.1016/j.ymssp.2019.106550) Crossref, ISI, Google Scholar - 37.
Pan S, Yang Q . 2010A survey on transfer learning. IEEE Trans. Knowl. Data Eng. 22, 1345–1359. (doi:10.1109/TKDE.2009.191) Crossref, ISI, Google Scholar - 38.
Long M, Wang J, Ding G, Sun J, Yu P . 2013Transfer feature learning with joint distribution adaptation. In 2013 IEEE Int. Conf. on Computer Vision, pp. 2200–2207. Google Scholar - 39.
Pan S, Tsang I, Kwok J, Yang Q . 2011Domain adaptation via transfer component analysis. IEEE Trans. Neural Netw. 22, 199–210. (doi:10.1109/TNN.2010.2091281) Crossref, PubMed, Google Scholar - 40.
Gretton A, Borgwardt K, Rasch M, Schöolkopf B, Smola A . 2012A kernel two-sample test. J. Mach. Learn. Res. 13, 723–773. ISI, Google Scholar - 41.
Schölkopf B, Smola A, Müller KR . 1998Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput. 10, 1299–1319. (doi:10.1162/089976698300017467) Crossref, ISI, Google Scholar
