Work

Data-Driven process parameter optimization of a tumbling self-piercing riveting machine via Bayesian optimization

Year: 2026
Type: article
Abstract: Tumbling self-piercing riveting (T-SPR) extends conventional self-piercing riveting by superimposing a controlled tumbling motion of the punch on its linear movement. While T-SPR improves the versatility of conventional SPR, it also increases the complexity and enlarges the space of adjustable process parameters. This work presents a Bayesian optimization framework for a data-driven optimization of these parameters. Two sequential single-objective optimization campaigns are conducted on a steel-aluminum material stack to optimize mechanical and geometric joint quality. The approach identifies parameter sets that substantially improve joining performance, achieving large reductions in joining force together with increases in tensile shear strength and energy absorption. The results demonstrate that Bayesian optimization enables sample-efficient exploration and optimization of the T-SPR parameter space.
Language: English
Open Access status: hybrid
Estimated APC paid: $3,290
OpenAlex ID: W7225307015
FWCI:
Cites: 18
Cited by:
Related to: 10