Karthigha, M (2026) Machine learning driven optimization of process parameters for dissimilar joints of Al6061 and A588K HSLA steel: an experimental approach. Cluster Computing, 29 (1). ISSN 1386-7857
Machine learning driven optimization of process parameters for dissimilar joints of Al6061 and A588K HSLA steel an experimental approach.pdf - Published Version
Download (10MB)
Abstract
Dissimilar metals of Al 6061 and A588 grade K high-strength low-alloy (HSLA) steel of 3 mm thickness were butts joined using laser and MIG welding processes with ER4043 filler metal. This study investigates the feasibility of joining Al alloy/steel. First, the preliminary investigation confirmed the bead appearance and mechanical and morphological characteristics of the selected laser power range of 4–12 kW and 85–100 A for MIG. Later, a machine learning model was developed to identify the best process parameters for both processes. Finally, a detailed investigation was performed with optimized process parameters confirming microstructure characteristics with scanning electron microscopy, X-ray diffraction, and fracture studies. The characterization results indicated that the joint strength is relatively based on intermetallic zone formation between molten metal and steel at high temperatures. Moreover, grain variance was revealed in the MIG and laser welding processes due to the difference in temperature gradient during the process. It also indicated that all welds displayed excellent surface finish and weld bead appearance, with plastically deformed steel particles dispersed at the Al alloy/steel interface. The greater number of alloying elements are dissipated in the fusion zone. Moreover, the higher other alloying elements are the rest of the two zones. The machine learning model showed strong correlation coefficients and aligned well with experimental values, demonstrating its accuracy and robustness. Overall, this study demonstrates that both laser and MIG welding, under optimized conditions, can achieve high-strength, high-quality Al6061-A588 HSLA steel joints.
| Item Type: | Article |
|---|---|
| Subjects: | C Computer Science and Engineering > Optimization Techniques C Computer Science and Engineering > Machine Learning |
| Divisions: | Computer Science and Engineering |
| Depositing User: | Dr Krishnamurthy V |
| Date Deposited: | 18 Dec 2025 05:03 |
| Last Modified: | 18 Dec 2025 05:12 |
| URI: | https://ir.psgitech.ac.in/id/eprint/1549 |
Dimensions
Dimensions