A simulation-based thermal process control method for multi-material laser-joining operations
- Mikel Ortiz(corresponding author),
- ,
- ,
- ,
- Christos Papaioannou,
- Harry Bikas
- ,
- ,
- ,
- University of Patras,
- Asociación de Investigación Metalúrgica del Noroeste,
- Údarás Industrial Estate
Open access
Abstract
The use of multi-material assemblies is becoming increasingly common in sectors such as aeronautics. In this context, laser joining process offers the possibility to obtain strong and reliable metal-composite bonds without the need of mechanical fasteners or adhesives but temperature control during the process is a critical aspect to obtain good-quality joints. Controlling the temperature at the multi-material interface is the key to obtain high-quality joints. However, this is a variable that cannot be monitored during the process. In this paper, a method for thermal control in laser joining operations is presented which is based on creating temperature synthetic data, from ANSYS® simulation models, both for process planning optimization and for the creation of AI-based regression models to predict the interface temperature over time. A case study involving the laser joining of aluminium and PEEK (Polyether Ether Ketone) components is used. The results demonstrate that the experimentally validated temperature synthetic data, enhance process optimization, reducing the need for extensive physical experiments. Moreover, the creation and use of regression models has been demonstrated a viable approach for predicting the temperature at the interface. However, the accuracy on temperature prediction depends on the type of sensor used to monitor the temperature.
Research Output Information
Output type
Original language
EnglishPages from-to (Number of pages)
Pages 1036-1050 (15 pages)Journal (Volume, Issue Number)
International Journal of Computer Integrated Manufacturing (Volume 39, Issue 7)Publication milestones
- Accepted/In press - 2025
- Published - 2026
Publication status
ISSN
0951-192XPublication IDs
- Scopus: 105007558293
