Force data-driven machine learning for defects in friction stir welding

Guan, Wei; Zhao, Yanhua; Liu, Yongchang; Kang, Su Sok; Wang, Dongpo; Cui, Lei

doi:10.1016/j.scriptamat.2022.114765

Cited by 28 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the two tools worked on the specimen surfaces, the one-step double-side FSW has more complex forces than forces on the one-side FSW. These forces significantly affect the joint quality, including defects and mechanical properties of welding joints [39,40]. These forces should be directed to channel the energy produced into the workpiece joining process.…”

Section: Apparatus Design On One-step Double-side Fswmentioning

confidence: 99%

Apparatus Design of One-Step Double-Side Friction Stir Welding for Aluminum Plates

Muhayat

Putri

Hendrato

et al. 2023

Designs

View full text Add to dashboard Cite

Aluminum alloys emerged as one of the materials used in manufacturing automotive car bodies due to their advantageous properties such as high strength-to-weight ratio, relatively low cost, high ductility, and high corrosion resistance. However, joining aluminum alloys using fusion welding poses serious problems due to the high solubility of hydrogen gas, which causes porosity in welding metal. Subsequently, solid-state welding, such as friction stir welding (FSW), has been considered a porosity-free aluminum joining method. However, the method has limitations, such as low flexibility and the need for a complex clamping system. It is particularly problematic when welding plates. It causes the welding process to be carried out twice on opposite sides, resulting in longer production times. This study designed and assembled a one-step double-side FSW apparatus to address this challenge and conducted welding trials with various welding parameters. During the welding trial, the upper and lower tool rotation varied at 900/900 rpm and 1500/1500 rpm. As a result, one-step double-side FSW was successfully used for welding 6 mm aluminum without any porosity defects. Faster tool rotation results in a wider heat-affected area and higher tensile strength. In addition, the hard test showed that the one-step double-side FSW process had a lower hardness compared to the hardness of the base metal.

show abstract

Section: Apparatus Design On One-step Double-side Fswmentioning

confidence: 99%

Apparatus Design of One-Step Double-Side Friction Stir Welding for Aluminum Plates

Muhayat

Putri

Hendrato

et al. 2023

Designs

View full text Add to dashboard Cite

show abstract

“…At this condition, severe tool wear is observed after the welding experiment is finished, especially the probe threads and three flats. It has been reported that severe probe wear results in an insufficient material flow around the rotating tool during the welding process, and a high compression force between the tool and the workpiece due to material jam at the retreating side leads to a high F y [24].…”

Section: Welding Force Characteristicsmentioning

confidence: 99%

Prediction of M–A Constituents and Impact Toughness in Stir Zone of X80 Pipeline Steel Friction Stir Welds

Wang

et al. 2022

Acta Metall. Sin. (Engl. Lett.)

Self Cite

View full text Add to dashboard Cite

This study explored a strategy for predicting the proportion of martensite-austenite (M-A) constituents and impact toughness of stir zone (SZ) on X80 pipeline steel joints welded by friction stir welding (FSW). It is found that the welding forces, including the traverse force (F x ), the lateral force (F y ) and the plunge force (F z ), are the key variables related to the change of welding parameters and influence remarkably the characteristics of M-A constituents and impact toughness of SZ. The impact toughness of SZ is commonly lower than that of the base material due to the formation of lath bainite and coarsening of austenite. The characteristics of M-A constituents in SZ are sensitive to the variation of welding parameters and respond well to the change of welding forces. The proportion of small island M-A constituents increases with the decrease in rotational speed and the increase in F z . The increase in the amount of island M-A constituents is beneficial to improve the impact toughness of SZ. Based on the above findings, a machine learning (ML) model for predicting the M-A constituents and impact toughness is constructed using the force features as the input data set. The force data-driven ML model can predict the M-A constituents and impact toughness precisely and exhibits higher accuracy than ML built with welding parameters. It is believed that the high accuracy is achieved because the force features include more details of FSW process, such as the heat generation, material flow, plastic deformation, and so on, which govern the microstructural evolution of SZ during FSW.

show abstract

“…In line with improving predictive models, Verma et al [18] also tackled the prediction of tensile behavior in FS-welded AA7039 using ML, reinforcing the critical role of ML in achieving predictive accuracy that can anticipate the tensile outcomes of welded materials. The study by Guan et al [19] suggests a direction toward force data-driven ML for identifying defects in FSW, an application that may revolutionize quality assurance by automating the detection of FSW flaws.…”

Section: Introductionmentioning

confidence: 99%

Optimization of 2024-T3 Aluminum Alloy Friction Stir Welding Using Random Forest, XGBoost, and MLP Machine Learning Techniques

Myśliwiec,

Kubit,

Szawara

2024

Materials

View full text Add to dashboard Cite

This study optimized friction stir welding (FSW) parameters for 1.6 mm thick 2024T3 aluminum alloy sheets. A 3 × 3 factorial design was employed to explore tool rotation speeds (1100 to 1300 rpm) and welding speeds (140 to 180 mm/min). Static tensile tests revealed the joints’ maximum strength at 87% relative to the base material. Hyperparameter optimization was conducted for machine learning (ML) models, including random forest and XGBoost, and multilayer perceptron artificial neural network (MLP-ANN) models, using grid search. Welding parameter optimization and extrapolation were then carried out, with final strength predictions analyzed using response surface methodology (RSM). The ML models achieved over 98% accuracy in parameter regression, demonstrating significant effectiveness in FSW process enhancement. Experimentally validated, optimized parameters resulted in an FSW joint efficiency of 93% relative to the base material. This outcome highlights the critical role of advanced analytical techniques in improving welding quality and efficiency.

show abstract

Force data-driven machine learning for defects in friction stir welding

Cited by 28 publications

References 28 publications

Apparatus Design of One-Step Double-Side Friction Stir Welding for Aluminum Plates

Apparatus Design of One-Step Double-Side Friction Stir Welding for Aluminum Plates

Prediction of M–A Constituents and Impact Toughness in Stir Zone of X80 Pipeline Steel Friction Stir Welds

Optimization of 2024-T3 Aluminum Alloy Friction Stir Welding Using Random Forest, XGBoost, and MLP Machine Learning Techniques

Contact Info

Product

Resources

About