Void Defect Prediction in Friction Stir Welding using Machine Learning Algorithms
DOI:
https://doi.org/10.65093/aci.v16.n3.2025.36Keywords:
friction stir welding (FSW), volumetric defects, hybrid ensemble learning, thermomechanical modelingAbstract
This study proposes an advanced predictive framework based on hybrid machine learning architectures to model the non-linear relationship between thermomechanical parameters and failure probability. Through rigorous analysis of heterogeneous experimental data, state-of-the-art ensemble models were evaluated and optimized, including Stacking, Weighted Voting, and a Super-Ensemble. Results validate the superiority of hybrid architectures, achieving a classification accuracy of 86.4% and an Area Under the ROC Curve (AUC) of 0.93, outperforming conventional base estimators. Feature importance analysis via Mean Decrease in Impurity corroborated the physical phenomenology of the process, identifying rotational and welding speeds as the governing factors of plastic flow (71% of explained variance). This work not only demonstrates the feasibility of in-silico defect detection but also establishes the algorithmic foundations for the development of Digital Twins and adaptive control systems within the Industry 4.0 context.
Downloads
References
Avcı, A., Kocakulak, M., Acır, N., Gunes, E. & Canyurt, O.E. (2024). A study on the monitoring of weld quality using xgboost with particle swarm optimization. Ain Shams Engineering Journal, 15 (4), 102486. https://doi.org/10.1016/j.asej.2024.102651
Baruah, A. (2023). Void formation process data in welding. https://www.kaggle.com/datasets/arindambaruah/ void-formation-process-data-in-welding
Baruah, A., Nath, T. & Gope, J. (2023). Optimised machine learning classification model to detect void forma-tions in friction stir welding. Materials Today: Proceedings, 80, 3648–3653. https://doi.org/10.1016/j.matpr.2023.03.386
Breiman, L. (2001). Random forests. Machine Learning, 45 (1), 5–32. https://doi.org/10.1023/A:1010933404324
Chen, C., Wang, W., Li, Y. & Song, D. (2024). Machine learning-based characterization of friction stir welding process: A review. Journal of Adhesion Science and Technology, pages 1–26. https://doi.org/10.1080/01694243.2024.2345163
Chen, T. & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 785–794. https://doi.org/10.1145/2939672.2939785
Chuenmee, N., Jite-Aree, T. & Phannachitta, P. (2025). Machine learning for predicting resistance spot weld quality using process parameters. Journal of Industrial Information Integration, 37, 100559. https://doi.org/10.1016/j.rineng.2024.103570
Du, Y., Li, W., Wei, Y. & Li, J. (2019). Conditions for void formation in friction stir welding from machine learning. npj Computational Materials, 5 (1), 73. https://doi.org/10.1038/s41524-019-0207-y
Friedman, J.H. (2001). Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29 (5), 1189–1232. https://doi.org/10.1214/aos/1013203451
Jayashree, P.K., Gowda, B.R., Kumar, A., and S, S. (2018). Optimization of tig welding parameters for 6061al alloy using Taguchi´s based design of experiments. Materials Today: Proceedings, 5(5, Part 2):13449–13455. https://doi.org/10.1016/j.matpr.2018.10.154
León, F., Rojas, L., Gonzalez, C., Hernández, B. & García, J. (2024). Convergence frontier analysis of sparse-matrix quasi-newton methods: Applications to rock blasting geomechanics. Avances en Ciencia e Ingeniería, 15 (2), 51–63. https://doi.org/10.65093/aci.v15.n2.2024.13
Martínez, Y., Rojas, L., Peña, A., Valenzuela, M., Hernández, B. & García, J. (2024). Fractional pinns: New frontiers for long-range memory in continuum models. Avances en Ciencia e Ingeniería, 15 (3), 93–110. https://doi.org/10.65093/aci.v15.n3.2024.10
Mishra, R.S. & Ma, Z.Y. (2005). Friction stir welding and processing. Materials Science and Engineering: R: Reports, 50 (1-2), 1–78. https://doi.org/10.1016/j.mser.2005.07.001
Myśliwiec, P., Węglowski, M.S. & Dymek, S. (2024). Optimization of 2024-t3 aluminum alloy friction stir welding parameters using machine learning models. Materials, 17 (8),1866. https://doi.org/10.3390/ma17071452
Nandan, R., DebRoy, T. & Bhadeshia, H.K.D.H. (2008). Recent advances in friction-stir welding – process, weldment structure and properties. Progress in Materials Science, 53 (6), 980–1023. https://doi.org/10.1016/j.pmatsci.2008.05.001
Rajakumar, S., Muralidharan, C. & Balasubramanian, V. (2011). Optimisation and sensitivity analysis of friction stir welding process and tool parameters for joining aa1100 aluminium alloy. International Journal of Materials and Product Technology, 40 (1/2), 153–173. https://doi.org/10.1504/IJMMP.2011.040442
Rojas, L., Hernández, B. & García, J. (2025). A systematic review of intelligent agents, language models, and recurrent neural networks in industrial maintenance: Driving value creation for the mining sector. International Journal of Intelligent Systems, 2025(1):Article 9953223. Open access review article. ttps://doi.org/10.1155/int/9953223
Rojas, L., León, F., Bazan, V. & Hernández, B. (2024). Aproximaciones basadas en distancias ergódicas y mezclado caótico en pilas de lixiviación: Ergodic distances and chaotic mixing approaches for heap leaching systems. Avances en Ciencia e Ingeniería, 15 (2), 37–50. https://doi.org/10.65093/aci.v15.n2.2024.12
Schmidt, H.B., Hattel, J.H. & Wert, J. (2006). An analytical model for the heat generation in friction stir welding. Modelling and Simulation in Materials Science and Engineering, 14 (2), 143–157. https://doi.org/10.1088/0965-0393/12/1/013
Shi, L. & Wu, C. (2023). Thermal-fluid-structure coupling analysis of void defect in friction stir welding. Inter-national Journal of Mechanical Sciences, 241, 107959. https://doi.org/10.1016/j.ijmecsci.2022.107969
Tapia, S., Aguilera, G., Rojas, L. & García, J. (2024). Predictive maintenance based on machine learning: A systematic literature review and perspectives in industry 4.0. Avances en Ciencia e Ingeniería, 15 (4), 63–93. https://doi.org/10.65093/aci.v15.n4.2024.3
Thomas, W.M., Nicholas, E.D., Needham, J.C., Murch, M.G., Temple-Smith, P. & Dawes, C.J. (1991). Friction-stir butt welding. International Patent Application No. PCT/GB92/02203 and GB Patent Application No. 9125978.8.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
