Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods

Laser welding of steel and hardmetal presents significant challenges due to their differing material properties. Improper laser welding parameters can result in unstable joints, ultimately leading to reduced mechanical strength of the weld. Therefore, defining an optimal process window is critical t...

Full description

Saved in:
Bibliographic Details
Main Authors: Mohammadhossein Norouzian, Mahan Khakpour, Marko Orosnjak, Atal Anil Kumar, Slawomir Kedziora
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Journal of Advanced Joining Processes
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2666330925000391
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850123763112214528
author Mohammadhossein Norouzian
Mahan Khakpour
Marko Orosnjak
Atal Anil Kumar
Slawomir Kedziora
author_facet Mohammadhossein Norouzian
Mahan Khakpour
Marko Orosnjak
Atal Anil Kumar
Slawomir Kedziora
author_sort Mohammadhossein Norouzian
collection DOAJ
description Laser welding of steel and hardmetal presents significant challenges due to their differing material properties. Improper laser welding parameters can result in unstable joints, ultimately leading to reduced mechanical strength of the weld. Therefore, defining an optimal process window is critical to ensuring weld quality. In addition, a continuous process monitoring method like High-Speed Imaging (HSI) is essential in real industrial applications to maintain stability and detect potential defects. Understanding plume dynamics helps identify the most important features of weld quality, but it also provides deeper insight into operational parameters that discriminate different weld types. Analysis of individual image plume frames from HSI reveals distinct statistical features that are identified as unique to each welding condition. Performing systematic feature selection using plume morphology, spatter generation and weld quality, we achieved>95 % leveraging Machine Learning (ML) classifiers. Particularly, Gradient Boosting Classifier (GBC), Linear Discriminant Analysis (LDA), Multinomial Logistic Regression (MNL-LR), Support Vector Machine (SVM), and Random Forest (RF), where the RF obtained >99 % classification accuracy of weld quality. The RF was then used in performing Recursive Feature Elimination (RFE), and with the robustness analysis, we managed to reduce the number of features from forty-nine to nine features while maintaining satisfactory performance (Accuracy = 0.981, F1-score = 0.961, AUROC = 0.997). The position of the weld plume, plume eccentricity and plume width are the most essential features that lead to the improvement of node purity and classification accuracy.
format Article
id doaj-art-6d453877b33e4032b000db70aec0954a
institution OA Journals
issn 2666-3309
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Journal of Advanced Joining Processes
spelling doaj-art-6d453877b33e4032b000db70aec0954a2025-08-20T02:34:31ZengElsevierJournal of Advanced Joining Processes2666-33092025-06-011110031810.1016/j.jajp.2025.100318Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methodsMohammadhossein Norouzian0Mahan Khakpour1Marko Orosnjak2Atal Anil Kumar3Slawomir Kedziora4University of Luxembourg, 6, rue Coudenhove Kalergi, L-1359, Luxembourg; Corresponding author.University of Luxembourg, 6, rue Coudenhove Kalergi, L-1359, LuxembourgUniversity of Luxembourg, 6, rue Coudenhove Kalergi, L-1359, LuxembourgUniversité de Lorraine, Arts et Métiers ParisTech, LCFC, Metz, F-57000 Metz, FranceUniversity of Luxembourg, 6, rue Coudenhove Kalergi, L-1359, LuxembourgLaser welding of steel and hardmetal presents significant challenges due to their differing material properties. Improper laser welding parameters can result in unstable joints, ultimately leading to reduced mechanical strength of the weld. Therefore, defining an optimal process window is critical to ensuring weld quality. In addition, a continuous process monitoring method like High-Speed Imaging (HSI) is essential in real industrial applications to maintain stability and detect potential defects. Understanding plume dynamics helps identify the most important features of weld quality, but it also provides deeper insight into operational parameters that discriminate different weld types. Analysis of individual image plume frames from HSI reveals distinct statistical features that are identified as unique to each welding condition. Performing systematic feature selection using plume morphology, spatter generation and weld quality, we achieved>95 % leveraging Machine Learning (ML) classifiers. Particularly, Gradient Boosting Classifier (GBC), Linear Discriminant Analysis (LDA), Multinomial Logistic Regression (MNL-LR), Support Vector Machine (SVM), and Random Forest (RF), where the RF obtained >99 % classification accuracy of weld quality. The RF was then used in performing Recursive Feature Elimination (RFE), and with the robustness analysis, we managed to reduce the number of features from forty-nine to nine features while maintaining satisfactory performance (Accuracy = 0.981, F1-score = 0.961, AUROC = 0.997). The position of the weld plume, plume eccentricity and plume width are the most essential features that lead to the improvement of node purity and classification accuracy.http://www.sciencedirect.com/science/article/pii/S2666330925000391Laser weldingHardmetalSteelWeld quality predictionMachine learningHigh-speed imaging
spellingShingle Mohammadhossein Norouzian
Mahan Khakpour
Marko Orosnjak
Atal Anil Kumar
Slawomir Kedziora
Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
Journal of Advanced Joining Processes
Laser welding
Hardmetal
Steel
Weld quality prediction
Machine learning
High-speed imaging
title Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
title_full Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
title_fullStr Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
title_full_unstemmed Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
title_short Prediction of weld quality in laser welding of hardmetal and steel using high-speed imaging and machine learning methods
title_sort prediction of weld quality in laser welding of hardmetal and steel using high speed imaging and machine learning methods
topic Laser welding
Hardmetal
Steel
Weld quality prediction
Machine learning
High-speed imaging
url http://www.sciencedirect.com/science/article/pii/S2666330925000391
work_keys_str_mv AT mohammadhosseinnorouzian predictionofweldqualityinlaserweldingofhardmetalandsteelusinghighspeedimagingandmachinelearningmethods
AT mahankhakpour predictionofweldqualityinlaserweldingofhardmetalandsteelusinghighspeedimagingandmachinelearningmethods
AT markoorosnjak predictionofweldqualityinlaserweldingofhardmetalandsteelusinghighspeedimagingandmachinelearningmethods
AT atalanilkumar predictionofweldqualityinlaserweldingofhardmetalandsteelusinghighspeedimagingandmachinelearningmethods
AT slawomirkedziora predictionofweldqualityinlaserweldingofhardmetalandsteelusinghighspeedimagingandmachinelearningmethods