Predicting the weld width from high-speed successive images of the weld zone using different machine learning algorithms during laser welding

Wang Cai; Jianzhuang Wang; Longchao Cao; Gaoyang Mi; Leshi Shu; Qi Zhou; Ping Jiang; Wang Cai; Jianzhuang Wang; Longchao Cao; Gaoyang Mi; Leshi Shu; Qi Zhou; Ping Jiang

doi:10.3934/mbe.2019278

Mathematical Biosciences and Engineering

2019, Volume 16, Issue 5: 5595-5612. doi: 10.3934/mbe.2019278

Previous Article Next Article

Research article Special Issues

Predicting the weld width from high-speed successive images of the weld zone using different machine learning algorithms during laser welding

1.
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China
2.
School of Materials Science and Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China
3.
School of Aerospace Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China

Received: 28 March 2019 Accepted: 12 June 2019 Published: 17 June 2019

The dynamic behavior of the keyhole and molten pool is associated with the quality of weld seam. In this study, an on-line visual monitoring system is devised to photograph the keyhole and molten pool during external magnetic field assisted laser welding on the AISI 2205 duplex stainless steel plates. Seven features are defined to describe the morphology of the keyhole and molten pool. Then, the principal component analysis (PCA) algorithm is applied to reduce the dimensions of these features to obtain different number of principal components (PCs). Three different machine learning algorithms, i.e. the back propagation neural network (BPNN), the radial based function neural network (RBFNN) and the support vector regression (SVR), are utilized to fit the relationship between the chosen PCs and the weld width. Finally, the global and local prediction accuracy of these three machine learning algorithms are compared under different number of PCs. Results illustrated that data dimensionality reduction is helpful to improve the modeling efficiency. Machine learning algorithms can be exploited to predict the weld quality during laser welding with high accuracy. Among them, the BPNN model performs best and SVR model performs better than RBFNN model in this research. This work aims to model the relation between the features in weld zone and the weld quality with different machine learning algorithms, and provides a guideline of model selection for laser welding on-line monitoring and a necessary foundation for realizing intelligent welding with advanced algorithm.
- laser welding,
- keyhole,
- molten pool,
- image processing,
- machine learning algorithm,
- weld width
Citation: Wang Cai, Jianzhuang Wang, Longchao Cao, Gaoyang Mi, Leshi Shu, Qi Zhou, Ping Jiang. Predicting the weld width from high-speed successive images of the weld zone using different machine learning algorithms during laser welding[J]. Mathematical Biosciences and Engineering, 2019, 16(5): 5595-5612. doi: 10.3934/mbe.2019278

Related Papers:

Abstract

The dynamic behavior of the keyhole and molten pool is associated with the quality of weld seam. In this study, an on-line visual monitoring system is devised to photograph the keyhole and molten pool during external magnetic field assisted laser welding on the AISI 2205 duplex stainless steel plates. Seven features are defined to describe the morphology of the keyhole and molten pool. Then, the principal component analysis (PCA) algorithm is applied to reduce the dimensions of these features to obtain different number of principal components (PCs). Three different machine learning algorithms, i.e. the back propagation neural network (BPNN), the radial based function neural network (RBFNN) and the support vector regression (SVR), are utilized to fit the relationship between the chosen PCs and the weld width. Finally, the global and local prediction accuracy of these three machine learning algorithms are compared under different number of PCs. Results illustrated that data dimensionality reduction is helpful to improve the modeling efficiency. Machine learning algorithms can be exploited to predict the weld quality during laser welding with high accuracy. Among them, the BPNN model performs best and SVR model performs better than RBFNN model in this research. This work aims to model the relation between the features in weld zone and the weld quality with different machine learning algorithms, and provides a guideline of model selection for laser welding on-line monitoring and a necessary foundation for realizing intelligent welding with advanced algorithm.

References

[1]	E. Assunção, L. Quintino and R. Miranda, Comparative study of laser welding in tailor blanks for the automotive industry, Int. J. Adv. Manuf. Tech., 49 (2010), 123–131.
[2]	X. Cao, M. Jahazi, J. Immarigeon, et al., A review of laser welding techniques for magnesium alloys,J. Mater. Process. Tech., 171 (2006), 188–204.
[3]	R. T. Yang and Z. W. Chen, A study on fiber laser lap welding of thin stainless steel, Int. J. Precis. Eng. Man., 14 (2013), 207–214.
[4]	J. Stavridis, A. Papacharalampopoulos and P. Stavropoulos, Quality assessment in laser welding: a critical review, Int. J. Adv. Manuf. Tech., 94 (2017), 1825–1847.
[5]	M. Jager and F. A. Hamprecht, Principal component imagery for the quality monitoring of dynamic laser welding processes, IEEE T. Ind. Electron., 56 (2008), 1307–1313.
[6]	L. E. Afanas'eva, I. A. Barabonova, P. O. Zorenko, et al., Laser welding in external electrical and magnetic fields, Weld. Int., 27 (2013), 545–547.
[7]	A. Schneider, V. Avilov, A. Gumenyuk, et al., Laser beam welding of aluminum alloys under the influence of an electromagnetic field, Phys. Procedia, 41 (2013), 4–11.
[8]	J. Volpp, Keyhole stability during laser welding-Part II: process pores and spatters, Prod. Eng. Res. Devel., 11 (2016), 9–18.
[9]	J. Stavridis, A. Papacharalampopoulos and P. Stavropoulos, A cognitive approach for quality assessment in laser welding, Procedia CIRP, 72 (2018), 1542–1547.
[10]	C. H. Kim and D. C. Ahn, Coaxial monitoring of keyhole during Yb:YAG laser welding, Opt. Laser Technol., 44 (2012), 1874–1880.
[11]	A. Matsunawa, J. D. Kim, N. Seto, et al., Dynamics of keyhole and molten pool in laser welding, J. Laser Appl., 10 (1998), 247–254.
[12]	G. Chen, M. Zhang, Z. Zhao, et al., Measurements of laser-induced plasma temperature field in deep penetration laser welding, Opt. Laser Technol., 45 (2013), 551–557.
[13]	J. Wang, C. Wang, X. Meng, et al., Study on the periodic oscillation of plasma/vapour induced during high power fibre laser penetration welding, Opt. Laser Technol., 44 (2012), 67–70.
[14]	C. Fan, F. Lv and S. Chen, Visual sensing and penetration control in aluminum alloy pulsed GTA welding, Int. J. Adv. Manuf. Tech., 42 (2009), 126–137.
[15]	R. Fabbro, S. Slimani, F. Coste, et al., Study of keyhole behaviour for full penetration Nd–Yag CW laser welding, J. Phys. D. Appl. Phys., 38 (2005), 1881.
[16]	K. Stefan, Process monitoring and control of laser beam welding: Measuring quantifiable data for improved processing results, Laser Technik J., 5 (2008), 41–43.
[17]	K. Stefan, Understanding the Laser Process new approaches for Process monitoring in laser materials Processing, Laser Technik J., 7, (2010), 49–52.
[18]	Z. Al-Sarraf and M. Lucas, A study of weld quality in ultrasonic spot welding of similar and dissimilar metals, J. Phys. Conference Series, 382 (2012), 012013.
[19]	L. Yang and I. C. Ume, Inspection of simulated weld penetration depth using laser-generated Lamb waves and wavelet signal processing, AIP Conference Proceedings. AIP, 1650 (2015), 1386–1391.
[20]	M. A. Maher, P. J. L. Webster, J. C. Chiao, et al., Coaxial real-time metrology and gas assisted laser micromachining: process development, stochastic behavior, and feedback control, Micromachining and Microfabrication Process Technology XV. International Society for Optics and Photonics, 7590 (2010), 759003.
[21]	P. J. L. Webster, G. L. Wright, Y. Ji, et al., Automatic laser welding and milling with in situ inline coherent imaging, Opt. Lett., 39 (2014), 6217–6220.
[22]	A. Khan, B. Baharudin, L. H. Lee, et al., A review of machine learning algorithms for text-documents classification, J. Adv. Inform. Tech., 1 (2010), 4–20.
[23]	P. Aivaliotis, A. Zampetis, G. Michalos, et al., A machine learning approach for visual recognition of complex parts in robotic manipulation, Procedia Manuf., 11 (2017), 423–430.
[24]	K. Kokkalis, G. Michalos, P. Aivaliotis, et al., An approach for implementing power and force limiting in sensorless industrial robots, Procedia CIRP, 76 (2018), 138–143.
[25]	X. Gao, C. Lan, D. You, et al., Weldment Nondestructive Testing Using Magneto-optical Imaging Induced by Alternating Magnetic Field,J. Nondestruct. Eval., 36 (2017), 55.
[26]	T. K. Xia, P. M. Hui and D. Stroud, Theory of faraday rotation in granular magnetic materials, J. Appl. Phys., 67 (1990), 2736–2741.
[27]	X. Wan, Y. Wang, D. Zhao, et al., Weld quality monitoring research in small scale resistance spot welding by dynamic resistance and neural network, Measurement, 99 (2017), 120–127.
[28]	M. Luo and Y. C. Shin, Estimation of keyhole geometry and prediction of welding defects during laser welding based on a vision system and a radial basis function neural network, Int. J. Adv. Manuf. Tech., 81 (2015), 263–276.
[29]	T. Wang, J. Chen, X. Gao, et al., Quality Monitoring for Laser Welding Based on High-Speed Photography and Support Vector Machine, Appl. Sci., 7 (2017), 299.
[30]	T. Wang, J. Chen, X. Gao, et al., Real-time monitoring for disk laser welding based on feature selection and SVM, Appl. Sci., 7, (2017), 884.
[31]	Z. Zhang, E. Kannatey-Asibu, S. Chen, et al., Online defect detection of Al alloy in arc welding based on feature extraction of arc spectroscopy signal, Int. J. Adv. Manuf. Tech., 79 (2015), 2067–2077.
[32]	J. Michalska and M. Sozańska, Qualitative and quantitative analysis of σ and χ phases in 2205 duplex stainless steel, Mater. Charact., 56 (2006), 355–362.
[33]	S. M. Holland, Principal components analysis (PCA), Department of Geology, University of Georgia, Athens,GA (2008), 30602–2501.
[34]	M. Buscema, Back propagation neural networks, Int. J. Addictions, 33 (1998), 233–270.
[35]	S. Chen, C. F. Cowan and P. M. Grant, Orthogonal least squares learning algorithm for radial basis function networks, IEEE T. Neural Networ., 2 (1991), 302–309.
[36]	H. Drucker, C. J. Burges, L. Kaufman, et al., Support vector regression machines, Adv. Neural Inform. Process. Sys., (1997), 155–161.
[37]	S. Altarazi, L. Hijazi and E. Kaiser, Process parameters optimization for multiple-inputs- multiple-outputs pulsed green laser welding via response surface methodology, 2016 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM). IEEE, (2016), 1041–1045.
[38]	C. J. Willmott, Some comments on the evaluation of model performance, B. Am. Meteorol. Soc., 63 (1982), 1309–1313.
[39]	J. Qian, J. Yi, Y. Cheng, et al., A sequential constraints updating approach for Kriging surrogate model-assisted engineering optimization design problem, Eng. Comput-Germany., (2019), 1–17.
[40]	Z. H. Han, C. Z. Xu, Z. Liang, et al., Efficient aerodynamic shape optimization using variable-fidelity surrogate models and multilevel computational grids, Chinese J. Aeronaut., (2019).

Reader Comments

Your name:*

Email:*
© 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)