scholarly journals Process Monitoring in Friction Stir Welding Using Convolutional Neural Networks

Metals ◽  
2021 ◽  
Vol 11 (4) ◽  
pp. 535
Author(s):  
Roman Hartl ◽  
Andreas Bachmann ◽  
Jan Bernd Habedank ◽  
Thomas Semm ◽  
Michael F. Zaeh
Keyword(s):  
Neural Networks ◽  
Friction Stir Welding ◽  
Convolutional Neural Networks ◽  
Process Monitoring ◽  
Surface Defects ◽  
Network Architectures ◽  
Process Data ◽  
Friction Stir ◽  
Measured Process

Preliminary studies have shown the superiority of convolutional neural networks (CNNs) compared to other network architectures for determining the surface quality of friction stir welds. In this paper, CNNs were employed to detect cavities inside friction stir welds by evaluating inline measured process data. The aim was to determine whether CNNs are suitable for identifying surface defects exclusively, or if the approach is transferable to internal weld defects. For this purpose, 120 welds were produced and examined by ultrasonic testing, which was the basis for labeling the data as “good” or “defective.” Different types of artificial neural network were tested for predicting the placement of the welds into the defined classes. It was found that the way of labeling the data is significant for the accuracy achievable. When the complete welds were uniformly labeled as “good” or “defective,” an accuracy of 98.5% was achieved by a CNN, which was a significant improvement compared to the state of the art. When the welds were labeled segment-wise, an accuracy of 79.2% was obtained by using a CNN, showing that a segment-wise prediction of the cavities is also possible. The results confirm that CNNs are well suited for process monitoring in friction stir welding and their application enables the identification of various defect types.

Prediction of the surface quality of friction stir welds by the analysis of process data using Artificial Neural Networks

2020 ◽  
Vol 234 (5) ◽  
pp. 732-751 ◽  
Author(s):  
R Hartl ◽  
B Praehofer ◽  
MF Zaeh
Keyword(s):  
Neural Networks ◽  
Friction Stir Welding ◽  
Surface Quality ◽  
Convolutional Neural Networks ◽  
Visual Inspection ◽  
Process Data ◽  
Friction Stir ◽  
Artificial Neural ◽  
Topography Analysis

Friction stir welding is an advanced joining technology that is particularly suitable for aluminum alloys. Various studies have shown that welding quality depends significantly on the welding speed and the rotational speed of the tool. It is frequently possible to detect an unsuitable setting of these parameters by examining the resulting surface defects, such as increased flash formation or surface galling. In this work, Artificial Neural Networks were used to analyze process data in friction stir welding and predict the resulting quality of the weld surface. For this purpose, nine different variables were recorded during friction stir welding of EN AW-6082 T6 sheets: the forces and accelerations in three spatial directions, the spindle torque, and temperatures at the tool shoulder and tool probe. In Case 1, the welds were assigned to the classes good and defective on the basis of a human visual inspection of the weld surface. In Case 2, the welds were categorized into the two classes on the basis of a surface topography analysis. Subsequently, three different major Artificial Neural Network architectures were tested for their ability to predict the surface quality: Feed Forward Fully Connected Neural Networks, Recurrent Neural Networks and Convolutional Neural Networks. The highest classification accuracy was achieved when Convolutional Neural Networks were used. Thus, the evaluation of the force signal transverse to the welding direction yielded the highest accuracy of 99.1% for the prediction of the result of the human visual inspection. The result achieved for the prediction of the topography analysis was an accuracy of 87.4% when the spindle torque was evaluated. Using all nine different process variables to predict the topography analysis, the accuracy could be improved slightly to 88.0%. The sampling rate of the spindle torque was varied between 40 Hz and 9600 Hz and no significant influence was determined. The findings show that Convolutional Neural Networks are well suited for the interpretation of friction stir welding process data and can be used to predict the resulting surface quality. In future work, the results are to be used to develop a parameter optimization method for friction stir welding.

scholarly journals Review on Friction Stir Processed TIG and Friction Stir Welded Dissimilar Alloy Joints

Metals ◽  
2020 ◽  
Vol 10 (1) ◽  
pp. 142 ◽  
Author(s):  
Sipokazi Mabuwa ◽  
Velaphi Msomi
Keyword(s):  
Friction Stir Welding ◽  
Friction Stir Processing ◽  
Aluminium Alloys ◽  
Processing Technique ◽  
Inert Gas ◽  
Tungsten Inert Gas Welding ◽  
Friction Stir ◽  
Dissimilar Alloys ◽  
The Status

There is an increase in reducing the weight of structures through the use of aluminium alloys in different industries like aerospace, automotive, etc. This growing interest will lead towards using dissimilar aluminium alloys which will require welding. Currently, tungsten inert gas welding and friction stir welding are the well-known techniques suitable for joining dissimilar aluminium alloys. The welding of dissimilar alloys has its own dynamics which impact on the quality of the weld. This then suggests that there should be a process which can be used to improve the welds of dissimilar alloys post their production. Friction stir processing is viewed as one of the techniques that could be used to improve the mechanical properties of a material. This paper reports on the status and the advancement of friction stir welding, tungsten inert gas welding and the friction stir processing technique. It further looks at the variation use of friction stir processing on tungsten inert gas and friction stir welded joints with the purpose of identifying the knowledge gap.

Powder-Bed Fusion Process Monitoring by Machine Vision With Hybrid Convolutional Neural Networks

2020 ◽  
Vol 16 (9) ◽  
pp. 5769-5779
Author(s):  
Yingjie Zhang ◽  
Hong Geok Soon ◽  
Dongsen Ye ◽  
Jerry Ying Hsi Fuh ◽  
Kunpeng Zhu
Keyword(s):  
Neural Networks ◽  
Machine Vision ◽  
Convolutional Neural Networks ◽  
Process Monitoring ◽  
Fusion Process ◽  
Powder Bed Fusion ◽  
Powder Bed

scholarly journals Friction Stir Welding of T-Joints: Experimental and Statistical Analysis

2019 ◽  
Vol 3 (2) ◽  
pp. 38 ◽  
Author(s):  
Ibrahim Sabry ◽  
Ahmed M. El-Kassas ◽  
Abdel-Hamid I. Mourad ◽  
Dinu Thomas Thekkuden ◽  
Jaber Abu Qudeiri
Keyword(s):  
Friction Stir Welding ◽  
Heat Affected Zone ◽  
Surface Defects ◽  
Welding Process ◽  
Travel Speed ◽  
Percentage Error ◽  
Nugget Zone ◽  
T Joints ◽  
Friction Stir ◽  
Full Factorial

T-welded joints are commonly seen in various industrial assemblies. An effort is made to check the applicability of friction stir welding for producing T-joints made of AA6063-T6 using a developed fixture. Quality T-joints were produced free from any surface defects. The effects of three parameters, such as the speed of rotation of the tool, axial force, and travel speed were analyzed. Correspondingly, mechanical characteristics such as tensile strength, hardness in three zones (thermal heat affected zone, heat affected zone, and nugget zone) and temperature distribution were measured. The full factorial analysis was performed with various combinations of parameters generated using factorial design and responses. Evident changes in the strength, hardness, and temperature profile were noticed for each combination of parameters. The three main parameters were significant in every response with p-values less than 0.05, indicating their importance in the friction stir welding process. Mathematical models developed for investigated responses were satisfactory with high R-sq and least percentage error.

Classification and identification of surface defects in friction stir welding: An image processing approach

2016 ◽  
Vol 22 ◽  
pp. 237-253 ◽  
Author(s):  
Ravi Ranjan ◽  
Aaquib Reza Khan ◽  
Chirag Parikh ◽  
Rahul Jain ◽  
Raju Prasad Mahto ◽  
...  
Keyword(s):  
Image Processing ◽  
Friction Stir Welding ◽  
Surface Defects ◽  
Friction Stir ◽  
Processing Approach

Numerical Simulation of Friction Stir Welding (FSW) Process Based on ABAQUS Environment

2016 ◽  
Vol 254 ◽  
pp. 272-277
Author(s):  
Monica Iordache ◽  
Claudiu Bădulescu ◽  
Eduard Niţu ◽  
Doina Iacomi
Keyword(s):  
Numerical Simulation ◽  
Friction Stir Welding ◽  
Finite Elements ◽  
Adaptive Mesh ◽  
Experimental Tests ◽  
Arbitrary Lagrangian Eulerian ◽  
Complex Issue ◽  
Friction Stir ◽  
Mechanical Phenomena

. Simulation of the FSW process is a complex issue, as it implies interactions between thermal and mechanical phenomena and the quality of the welding depends on many factors. In order to reduce the time of the experimental tests, which can be long and expensive, numerical simulation of the FSW process has been tried during the last ten years. However, there still remain aspects that cannot be completely simulated. In this paper the authors present the steps of the numerical simulation using the finite elements method, in order to evaluate the boundary conditions of the model and the geometry of the tools by using the Arbitrary Lagrangian Eulerian (ALE) adaptive mesh controls.

Sign in / Sign up

Export Citation Format

Share Document