Machine learning as a comparative tool to determine the relevance of signal features in laser welding

Knaak, Christian; Thombansen, Ulrich; Abels, Peter; Kröger, Moritz

doi:10.1016/j.procir.2018.08.073

Cited by 33 publications

(12 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…32 In laser welding, real-time process information has been found essential for quality assurance and process control. 33 Most of the literature's works focus on quality assurance, for example., porosity absence, the extension of the heat-affected zone (HAZ), and gap size; however, hardness is seldom explored. Singh et al 34 employed neural networks and genetic algorithms to model the hardness profile in plain carbon steel (DC05) laser welded.…”

Section: Introductionmentioning

confidence: 99%

Vibration-assisted laser welding: Frequency and amplitude vibration effect on the microhardness of A36 steel

Vidal

Barrionuevo

Ramos‐Grez

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part L:

View full text Add to dashboard Cite

Laser welding is one of the most versatile welding techniques for joining advanced materials. However, thermal cycles induced around weld lines generate residual stresses and inhomogeneous plastic deformation during the welding process. Vibratory weld conditioning is an alternative to enhance the mechanical properties of metal sheets during casting and welding. This study aims to analyze the effects of vibration on the microhardness of the resolidified melt zone after a laser weldment. Two 1 mm thick A36 steel sheets were welded in lap configuration using a CO2 laser, with varying vibration amplitudes up to 0.56 mm peak-to-peak and frequencies up to 1034 Hz. We confirmed that mechanical vibrations during the welding process alter the final properties of the joint compared to non-vibrated samples. Based on the experimental results, vibration can either increase or decrease the bead's final microhardness in a range of [−5%, 7%] by controlling the frequency and vibration amplitude. Furthermore, average microhardness was predicted by applying several machine learning regressors, showing an accuracy of ±3%, denoting the usefulness of the random forest and AdaBoost algorithms provide.

show abstract

Section: Introductionmentioning

confidence: 99%

Vibration-assisted laser welding: Frequency and amplitude vibration effect on the microhardness of A36 steel

Vidal

Barrionuevo

Ramos‐Grez

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part L:

View full text Add to dashboard Cite

show abstract

“…In the manufacturing domain, such data-driven approaches have been extensively studied in the past and are based on autoregressive (AR) models, cluster analysis, fuzzy set theory or on supervised learning algorithms such as multivariate regression, multi-layer perceptron and decision trees, as well as k-nearest neighbors [ 14 , 15 ]. Therefore, recent development led to advanced process monitoring systems which integrate machine learning techniques for process control and prediction of critical defects [ 16 , 17 ]. An advantage of data-driven methods is that it is not necessary to explicitly model the physical behavior of the system in order to build a statistical model.…”

Section: Introductionmentioning

confidence: 99%

A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards

Knaak

Eßen

Kröger

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

In modern production environments, advanced and intelligent process monitoring strategies are required to enable an unambiguous diagnosis of the process situation and thus of the final component quality. In addition, the ability to recognize the current state of product quality in real-time is an important prerequisite for autonomous and self-improving manufacturing systems. To address these needs, this study investigates a novel ensemble deep learning architecture based on convolutional neural networks (CNN), gated recurrent units (GRU) combined with high-performance classification algorithms such as k-nearest neighbors (kNN) and support vector machines (SVM). The architecture uses spatio-temporal features extracted from infrared image sequences to locate critical welding defects including lack of fusion (false friends), sagging, lack of penetration, and geometric deviations of the weld seam. In order to evaluate the proposed architecture, this study investigates a comprehensive scheme based on classical machine learning methods using manual feature extraction and state-of-the-art deep learning algorithms. Optimal hyperparameters for each algorithm are determined by an extensive grid search. Additional work is conducted to investigate the significance of various geometrical, statistical and spatio-temporal features extracted from the keyhole and weld pool regions. The proposed method is finally validated on previously unknown welding trials, achieving the highest detection rates and the most robust weld defect recognition among all classification methods investigated in this work. Ultimately, the ensemble deep neural network is implemented and optimized to operate on low-power embedded computing devices with low latency (1.1 ms), demonstrating sufficient performance for real-time applications.

show abstract

“…26 Examples, for instance in other fields of welding technologies, are given for process control 27 or quality assurance. 28 An ANN is a collection of nodes, hierarchically stacked by layers (see Figure 3), defined over connection weights and biases (see Figure 4). A weight contains a value by which a signal, floating through the node coming from the predecessor neuron, is amplified, minimized or conversed.…”

Section: Introductionmentioning

confidence: 99%

“…26 Examples, for instance in other fields of welding technologies, are given for process control 27 or quality assurance. 28…”

Section: Introductionmentioning

confidence: 99%

In-situ monitoring of hybrid friction diffusion bonded EN AW 1050/EN CW 004A lap joints using artificial neural nets

Kohler

Schiele

Glaser

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part L:

View full text Add to dashboard Cite

In this work, a dissimilar copper/aluminum lap joint was generated by force-controlled hybrid friction diffusion bonding setup (HFDB). During the welding process, the appearing torque, the welding force as well as the plunge depth are recorded over time. Due to the force-controlled process, tool wear and the use of different materials, the resulting data series varies significantly, which makes quality assurance according to classical methods very difficult. Therefore, a Convolutional Neural Network was developed which allows the evaluation of the recorded process data. In this study, data from sound welds as well as data from samples with weld defects were considered. In addition to the different welding qualities, deviations from the ideal conditions due to tool wear and the use of different alloys were also considered. The validity of the developed approach is determined by cross validation during the training process and different amounts of training data. With an accuracy of 88.5%, the approach of using Convolutional Neural Network has proven to be a suitable tool for monitoring the processes.

show abstract

Machine learning as a comparative tool to determine the relevance of signal features in laser welding

Cited by 33 publications

References 3 publications

Vibration-assisted laser welding: Frequency and amplitude vibration effect on the microhardness of A36 steel

Vibration-assisted laser welding: Frequency and amplitude vibration effect on the microhardness of A36 steel

A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards

In-situ monitoring of hybrid friction diffusion bonded EN AW 1050/EN CW 004A lap joints using artificial neural nets

Contact Info

Product

Resources

About