Machine Learning Techniques Comparison Applied to Structural Health Monitoring in Pipes

Oliveira, M. Post-Graduation Program in Mining, Metallurgy and Materials Engineering (PPGEM), Federal University of Rio Grande do Sul (UFRGS), Porto Alegre- Brazil, oliwer.marlon@gmail.com

Pazini, M. V. L. Post-Graduation Program in Mining, Metallurgy and Materials Engineering (PPGEM), Federal University of Rio Grande do Sul (UFRGS), Porto Alegre- Brazil, marcuspazini@gmail.com

Corrêa, L. A. Post-Graduation Program in Mining, Metallurgy and Materials Engineering (PPGEM), Federal University of Rio Grande do Sul (UFRGS), Porto Alegre- Brazil, labcorrea@gmail.com

Clarke, T. G. R. Post-Graduation Program in Mining, Metallurgy and Materials Engineering (PPGEM), Federal University of Rio Grande do Sul (UFRGS), Porto Alegre- Brazil, thomas.grclarke@gmail.com

Abstract

Structural Health Monitoring systems using the ultrasonic guided wave method are able to identify, locate and estimate the severity of defects resulting from corrosive processes in structures. Such systems, when constituted of sensors permanently coupled or inserted in the structure and electronics of acquisition and transmission of data, are close to the concept of intelligent structures, which are those that have the ability to self-diagnose. One of the biggest problems for this approach is the sheer volume of data generated. The present work aims the elaboration of a methodology based on machine learning algorithms for processing data obtained from such systems. A synthetic database was produced for feed several algorithms such as kNN, CNN, MLP and SVM. A significantly high success rate was observed for the identification of the damaged structure, i.e., the method presented in this work proved to be very promising and at the moment a prototype is installed in a refinery, being tested under real operating conditions.

Keywords:

Machine Learning

SHM

Guided Wave inspection

Pipeline inspection.

References

[1] Christopher M Bishop. Pattern recognition and machine learning. springer, 2006.

[2] G. Shakhnarovich, P. Viola, and T. Darrell. Fast pose estimation with parameter-sensitive hashing. Proc. Ninth IEEE Int. Conf. Comput. Vis., (April):0–11, 2003. doi:10.1109/ICCV.2003.1238424.

[3] LA Bull, PA Gardner, J Gosliga, TJ Rogers, N Dervilis, EJ Cross, E Papatheou, AE Maguire, C Campos, and K Worden. Foundations of population-based shm, part i: Homogeneous pop ulations and forms. Mechanical Systems and Signal Processing, 148:107141, 2020.

[4] Marlon Oliveira, Houssem Chatbri, Ylva Ferstl, Suzanne Little, Noel E. O’Connor, and Alistair Sutherland. Irish sign language recognition using principal component analysis and convolutional neu ral networks. In 2017 International Conference on Digital Image Computing: Techniques and Ap plications (DICTA), pages 1–8. IEEE, Nov 2017. doi:10.1109/DICTA.2016.7796984.

[5] Marlon Oliveira, Houssem Chatbri, Suzanne Little, Noel E. O’Connor, and Alistair Sutherland. A com parison between end-to-end approaches and feature extraction based approaches for sign language recog nition. In International Conference on Image and Vi sion Computing New Zealand (IVCNZ), 2017, pages 1–6. IEEE, 2017.

[6] Majdi Flah, Itzel Nunez, Wassim Ben Chaabene, and Moncef L Nehdi. Machine learning algorithms in civil structural health monitoring: a systematic review. Archives of Computational Methods in Engi neering, pages 1–23, 2020.

[7] Thomas Clarke, Peter Cawley, Paul David Wilcox, and Anthony John Croxford. Evaluation of the dam age detection capability of a sparse-array guided wave SHM system applied to a complex structure under varying thermal conditions. IEEE Transac tions on Ultrasonics, Ferroelectrics, and Frequency Control, 56(12):2666–2678, 2009. ISSN 08853010. doi:10.1109/TUFFC.2009.1357.

[8] Ajay Raghavan. Guided-wave structural health monitoring. PhD thesis, 2007.

[9] Mira Mitra and S Gopalakrishnan. Guided wave based structural health monitoring: A review. Smart Materials and Structures, 25(5):053001, 2016.

[10] Christoph Kralovec and Martin Schagerl. Review of structural health monitoring methods regarding a multi-sensor approach for damage assessment of metal and composite structures. Sensors, 20(3):826, 2020.

[11] Charles R Farrar and Keith Worden. Structural health monitoring: a machine learning perspective. John Wiley & Sons, 2012.

[12] Yujie Ying, Joel Harley, James H. Garrett, Yuan wei Jin Jr., Irving J. Oppenheim, Jun Shi, and Lucio Soibelman. Applications of Machine Learning in Pipeline Monitoring. International Workshop on Computing in Civil Engineering, (2011):194–201, 2011.

[13] Joseph Melville, K. Supreet Alguri, Chris Deemer, and Joel B. Harley. Structural damage detection using deep learning of ultrasonic guided waves. AIP Conference Proceedings, 1949(April), 2018. ISSN 15517616. doi:10.1063/1.5031651.

[14] Arnaud Mignan and Marco Broccardo. One neu ron versus deep learning in aftershock prediction. Nature, 574(7776):E1–E3, 2019.

[15] Arnaud Mignan and Marco Broccardo. A deeper look into ‘deep learning of aftershock patterns fol lowing large earthquakes’: Illustrating first princi ples in neural network physical interpretability. In International Work-Conference on Artificial Neural Networks, pages 3–14. Springer, 2019.

[16] Debarshi Sen, Amirali Aghazadeh, Ali Mousavi, Satish Nagarajaiah, Richard Baraniuk, and Anand Dabak. Data-driven semi-supervised and supervised learning algorithms for health monitoring of pipes. MechanicalSystems andSignalProcessing, 131:524 537, 2019.

[17] Chang Liu, Jacob Dobson, and Peter Cawley. Effi cient generation of receiver operating characteristics for the evaluation of damage detection in practical structural health monitoring applications. Proceed ings of the Royal Society A: Mathematical, Physi cal and Engineering Sciences, 473(2199):20160736, 2017.

[18] Pi ceramic. https://www.piceramic.com/en/. Accessed: 2020-11-04.

[19] Ricardo C Jacques, Henrique H de Oliveira, Rafael WF dos Santos, and Thomas GR Clarke. De sign and in situ validation of a guided wave sys tem for corrosion monitoring in coated buried steel pipes. Journal of Nondestructive Evaluation, 38(3): 65, 2019.

[20] Mike JS Lowe, David N Alleyne, and Peter Caw ley. Defect detection in pipes using guided waves. Ultrasonics, 36(1-5):147–154, 1998.

[21] MHSSiqueira, CEN Gatts, RR Da Silva, and JMA Rebello. The use of ultrasonic guided waves and wavelets analysis in pipe inspection. Ultrasonics, 41 (10):785–797, 2004.

[22] RC Jacques, HTH De Oliveira, RWF Dos Santos, and TGR Clarke. Design of an emat guided wave collar for coated riser inspection (march 2020). IEEE Sensors Journal, 2020.

[23] Yihui Da, Bin Wang, DZ Liu, and Zhenghua Qian. An analytical approach to reconstruction of axisym metric defects in pipelines using t (0, 1) guided waves. Applied Mathematics and Mechanics, 41 (10):1479–1492, 2020.

[24] Paul D Wilcox. Guided-wave array methods. Ency clopedia of structural health monitoring, 2009.

[25] Vincentius Ewald, Roger M. Groves, and Rinze Benedictus. DeepSHM: adeeplearningapproachfor structural health monitoring based on guided Lamb wave technique. page 19, 2019. ISSN 1996756X. doi:10.1117/12.2506794.

[26] Jesper Sören Dramsch. Machine learning in 4d seismic data analysis: Deep neural networks in geo physics. 2019.

[27] Armand Wirgin. The inverse crime. arXiv preprint math-ph/0401050, 2004.

[28] Jari Kaipio and Erkki Somersalo. Statistical inverse problems: discretization, model reduction and in verse crimes. Journal of computational and applied mathematics, 198(2):493–504, 2007.

[29] Carlos Eduardo Chávez, Felipe Alonzo-Atienza, and Diego Alvarez. Avoiding the inverse crime in the inverse problem of electrocardiography: estimating the shape and location of cardiac ischemia. In Com puting in Cardiology 2013, pages 687–690. IEEE, 2013.

[30] Mahjoub El Mountassir, Slah Yaacoubi, and Fethi Dahmene. Reducing false alarms in guided waves structural health monitoring of pipelines: Review synthesis and debate. International Journal of Pres sure Vessels and Piping, page 104210, 2020.

[31] Panos Z Marmarelis and Vasilis Z Marmarelis. The white-noise method in system identification. In Analysis of physiological systems, pages 131–180. Springer, 1978.

[32] Lennart Ljung. System identification. Wiley ency clopedia of electrical and electronics engineering, pages 1–19, 1999.

[33] Ronald Newbold Bracewell and Ronald N Bracewell. The Fourier transform and its applications, volume 31999. McGraw-Hill New York, 1986.

[34] Simon Haykin and Barry Van Veen. Signals and systems. John Wiley & Sons, 2007.

[35] Matteo Frigo and Steven G. Johnson. The design and implementation of FFTW3. Proceedings of the IEEE, 93(2):216–231, 2005. Special issue on “Program Generation, Optimization, and Platform Adaptation”.

[36] Henrique Tormen Haan de Oliveira. Projeto de um colar de ondas guiadas para aplicação em tubulação enterrada. Master’s thesis, Universidade Federal do Rio Grande do Sul, 2017.

[37] R Naoum, HH Owaied, and S Joudeh. Development of a new Arabic sign language recognition using k-nearest neighbor algorithm. J. Emerg. Trends Comput. Inf. Sci., 3(8):1173–1178, 2012. URL http://isite.persiangig.com/Files/KNN/Arabic{_}N aoum.pdf.

[38] Simon Tong and Daphne Koller. Support Vec tor Machine Active Learning with Applications to Text Classification. J. Res., pages 45–66, 2001. Mach. Learn. ISSN 15324435. doi:10.1162/153244302760185243.

[39] Giorgio Licciardi, Ruggero Giuseppe Avezzano, Fabio Del Frate, Giovanni Schiavon, and Jocelyn Chanussot. A novel approach to polarimetric SAR data processing based on Nonlinear PCA. Pattern Recognit., 47(5):1953–1967, 2014. ISSN 00313203. doi:10.1016/j.patcog.2013.11.009.

[40] M.W Gardner and S.R Dorling. Artificial neu ral networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. At mos. Environ., 32(14-15):2627–2636, 1998. ISSN 13522310. doi:10.1016/S1352-2310(97)00447-0. URLhttp://www.sciencedirect.com/science/article/ pii/S1352231097004470.

[41] Jesper Sören. General rights Machine Learning in 4DSeismic Data Analysis: Deep Neural Networks in Geophysics. 2020. URL https://www.fysik.dtu.dk.

[42] WZhao, R. Chellappa, and S. Krishnaswamy. Dis criminant analysis of principal components for face recognition. Face Recognit. From Theory to Appl., pages 73–85, 1998. ISSN 1543-5474. doi:10.1109/AFGR.1998.670971. URL http://ieeexp lore.ieee.org/xpls/abs{_}all.jsp?arnumber=670971.

[43] Oya Aran and Lale Akarun. A multi-class classifica tion strategy for Fisher scores: Application to signer independent sign language recognition. Pattern Recognit., 43(5):1776–1788, 2010. ISSN 00313203. doi:10.1016/j.patcog.2009.12.002.

[44] Jerome Friedman, Trevor Hastie, and Robert Tibshi rani. The elements of statistical learning, volume 1. Springer series in statistics New York, 2001.