PARAMETRICALLY ROBUST NEURAL NETWORK ESTIMATOR OF ELECTRODE POSITION IN AUTOMATIC AND ROBOTIC WELDING WITH ARC OSCILLATIONS
DOI:
https://doi.org/10.35546/kntu2078-4481.2024.4.13Keywords:
neural network, estimator, parametric robustness, extremum, electrode position, arc oscillations, welding robotAbstract
The development of automatic and robotic welding systems is largely based on the use of tools to adapt the path of the welding torch. Visual sensors are currently the most commonly used means of such adaptation. However, when welding with arc oscillations, to adapt the welding torch path to the actual place of joining parts, it is promising to estimate the position of the electrode relative to the welding line based on the welding arc current. Significant advantages of the estimators are the absence of additional equipment on the welding torch, as well as the combination of welding and measurement points. To detect extremum in adaptive welding systems with search oscillations, the synchronous detection method is used. However, the accuracy of the electrode position relative to the weld seam obtained by a synchronous detector is low due to the influence of instability of the welding circuit parameters. Therefore, the task of constructing an electrode position estimator for welding with arc oscillations that will have high accuracy and be insensitive to parametric changes is currently relevant. The aim of this work is to consider a new approach based on the use of an artificial neural network to solving the problem of estimating the electrode position relative to the joint line of parts in welding with arc oscillations, while ensuring robustness to changes in the parameters of the welding circuit. To construct an estimator of the electrode deviation from the joint line in robotic welding with arc oscillations, a twolayer neural network with a direct signal transmission architecture of the 20-13-2 type was used. The neural network estimator, based on the welding current signal, determines the relative position of the electrode twice during the arc oscillation period. Using mathematical modelling, it is shown that the processes of estimating the electrode deviation from the weld line are characterized by a sufficiently high accuracy and robustness to changes in the parameters of the welding circuit.
References
Pires J. N., Loureiro A., Bölmsjo G. Welding Robots, Springer-Verlag, 2006. DOI: https://doi.org/10.1007/1-84628-191-1.
Lin L., Bingqiang L., Yanbiao Z. Study on seam tracking system based on stripe type laser sensor and welding robot. Chin J Lasers, 42, 5, 2015, pp. 1-8.
Pryymak B., Korol S., Ostroverkhov M. Design of a digital following system of welding robot with a visual sensor. Proc. of the IEEE 19th Intern. Conf. on Smart Technologies “EUROCON-2021”, Lviv, Ukraine, July 6–8, 2021, pp. 66-70. DOI: 10.1109/eurocon52738.2021.9535643.
Приймак Б.І. Синтез та дослідження алгоритму керування ланкою зварювального робота з органом технічного зору. Вісник Херсонського національного технічного університету, № 4 (83), 2022. С. 29-36. DOI: https://doi.org/10.35546/kntu2078-4481.2022.4.3
W. Huang and R. Kovacevic, “Development of a real-time laser-based machine vision system to monitor and control welding processes”, The International Journal of Advanced Manufacturing Technology, vol. 63, no. 1-4, 2012, pp. 235-248.
Yang L., Liu Y., Peng J. Advances techniques of the structured light sensing in intelligent welding robots: a review. The International Journal of Advanced Manufacturing Technology, vol. 110. no. 3, 2020, pp.1027-1046. DOI: https://doi.org/10.1007/s00170-020-05524-2
Цибулькін Г.А. Дугові сенсорні системи для зварювальних роботів, Сталь, Київ, 2011, 145 с.
Singh, P. K., Patel, D., Prasad, S. B. Development of vibratory welding technique and tensile properties investigation of shielded metal arc welded joints. Indian Journal of Science and Technology, Vol. 9(35), 2016, pp. 1–6. DOI: doi: 10.17485/ijst/2016/v9i35/92846.
Luo, Y., Zhang, Z.L., Zhou, C.F. et al. Effect of oscillation parameters of narrow groove MAG welding on weld formation. J. Hebei Univ. Sci. Technol., 38, 6. 2017. DOI: https://doi.org/10.3390/app13095519.
Цибулькін Г.А. Адаптивне керування у дуговому зварюванні, Сталь, Київ, 2014, 171 с.
Pryymak B., Zhelinskyi M., Ostroverkhov M., Khalimovskyy O. Neural network based estimator of the electrode deviation in robotic welding with arc oscillations. CEUR Workshop Proceedings, 2023, 3628, pp. 425-432. Режим доступу: https://ceur-ws.org/Vol-3628/short16.pdf
