Comparative Evaluation of Machine Learning Approaches for Predicting Optimal Parameters in Cold Metal Transfer Welding of Aluminum 8000 Series
Keywords:
Cold Metal Transfer (CMT), Aluminum Alloys, Welding Parameters, Machine Learning, Random Forest, GPR, KNN, UTS PredictionAbstract
Cold Metal Transfer (CMT) welding is an advanced gas metal arc welding (GMAW) technique characterized by its low heat input, spatter-free arc, and precise control over metal deposition. Unlike conventional MIG/MAG welding, CMT separates the wire feeding and current control systems, enabling controlled short-circuit transfer, making it ideal for joining thin and dissimilar materials such as aluminum and magnesium alloys. This study explores the application of machine learning (ML) models—Random Forest, K-Nearest Neighbors (KNN), and Gaussian Process Regression (GPR)—to predict optimal CMT welding parameters (voltage, current, wire feed rate) based on the chemical composition of aluminum alloys. The goal is to achieve target Ultimate Tensile Strength (UTS) while minimizing experimental trials. A benchmark datapoint from aluminum 8011 was used to evaluate model accuracy, and all models were assessed using performance metrics such as Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). Results showed that GPR and Random Forest provided highly accurate, interpretable predictions. This work contributes to the development of a data-driven, inverse modeling framework for smart welding systems, enabling rapid parameter optimization across a wide range of aluminum alloys.
Cold Metal Transfer (CMT) welding is an advanced gas metal arc welding (GMAW) technique characterized by its low heat input, spatter-free arc, and precise control over metal deposition. Unlike conventional MIG/MAG welding, CMT separates the wire feeding and current control systems, enabling controlled short-circuit transfer, making it ideal for joining thin and dissimilar materials such as aluminum and magnesium alloys. This study explores the application of machine learning (ML) models—Random Forest, K-Nearest Neighbors (KNN), and Gaussian Process Regression (GPR)—to predict optimal CMT welding parameters (voltage, current, wire feed rate) based on the chemical composition of aluminum alloys. The goal is to achieve target Ultimate Tensile Strength (UTS) while minimizing experimental trials. A benchmark datapoint from aluminum 8011 was used to evaluate model accuracy, and all models were assessed using performance metrics such as Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). Results showed that GPR and Random Forest provided highly accurate, interpretable predictions. This work contributes to the development of a data-driven, inverse modeling framework for smart welding systems, enabling rapid parameter optimization across a wide range of aluminum alloys.
Downloads
Metrics
References
Fuse, K., Venkata, P., Reddy, R. M., & Bandhu, D. (2025). Machine learning classification approach for predicting tensile strength in aluminium alloy during friction stir welding. International Journal on Interactive Design and Manufacturing (IJIDeM), 19(1), 639-643
Mishra, A., Sefene, E. M., & Tsegaw, A. A. (2021). Process parameter optimization of friction stir welding on 6061AA using supervised machine learning regression-based algorithms. arXiv preprint arXiv:2109.00570.
Kalita, K., & Kalita, D. (2024). Enhancing welding quality through predictive modelling-Insights from machine learning techniques. MM Science Journal, December 2024.
Mishra, A., Sefene, E.M., Nidigonda, G., & Tsegaw, A.A. (2022). Performance evaluation of machine learning-based algorithm and Taguchi algorithm for the determination of the hardness value of the friction stir welded AA 6262 joints at a nugget zone. arXiv preprint arXiv:2203.11649.
Dorbane, A., Harrou, F., Anghel, D. C., & Sun, Y. (2024). Machine learning prediction of aluminum alloy stress–strain curves at variable temperatures with failure analysis. Journal of Failure Analysis and Prevention, 24(1), 229-244.
Mishra, A., Al-Sabur, R., & Jassim, A. K. (2022). Machine Learning Algorithms for Prediction of Penetration Depth and Geometrical Analysis of Weld in Friction Stir Spot Welding Process. arXiv preprint arXiv:2201.09725.
Mahmudi, E., & Farhangi, H. (2010). The influence of welding parameters on tensile behavior of friction stir welded Al 2024-t4 joints. Advanced Materials Research, 83, 439-448.
Rajakumar, S., Muralidharan, C., & Balasubramanian, V. (2011). Predicting tensile strength, hardness and corrosion rate of friction stir welded AA6061-T6 aluminium alloy joints. Materials & Design, 32(5), 2878-2890.
Maneiah, D., Mishra, D., Rao, K. P., & Raju, K. B. (2020). Process parameters optimization of friction stir welding for optimum tensile strength in Al 6061-T6 alloy butt welded joints. Materials Today: Proceedings, 27, 904-908.
Mishra, A. (2024). Biomimetic machine learning approach for prediction of mechanical properties of additive friction stir deposited aluminum alloys based walled structures. arXiv preprint arXiv:2408.05237.
S Sarsilmaz, F. and Kavuran, G., 2021. Prediction of the optimal FSW process parameters for joints using machine learning techniques. Materials Testing, 63(11), pp.1104-1111.
Mishra, A., & Morisetty, R. (2022). Determination of the ultimate tensile strength (UTS) of friction stir welded similar AA6061 joints by using supervised machine learning based algorithms. Materials Today.
Verma, S., & Natarajan, S. (2020). Prediction of tensile behavior of FS welded AA7039 using machine learning approaches. Materials Today: Proceedings, 33, 1234-1240.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
