Accelerating Process Development of Aluminum-Intensive Dissimilar Friction Self-Piercing Riveting via Regression Analysis

Modern automotive structures increasingly rely on high-strength wrought and cast aluminum (Al) alloys to achieve lightweighting and improve structural performance. However, conventional joining techniques face significant challenge due to limited ductility at room temperature and high susceptibility to solidification cracking and porosity. Friction self-piercing riveting (F-SPR) addresses these limitations by using a rotating rivet to generate frictional heat, locally softening the surrounding materials and enabling crack free joints. For Al-intensive dissimilar and multi-layer stacks, a multi-step F-SPR process is required to accommodate complex joining conditions but introduces a high-dimensional parameter space that complicates process development. In this work, regression analysis was employed to effectively identify dominant process parameters and establish process–structure–property correlations with reduced experimental trials. The down-selected process parameters demonstrated sound joint and mechanical performance. This regression-based framework enables future integration of advanced AI/ML models to accelerate process development and scale-up for automotive applications.

This project is funded by the U.S. Department of Energy, Vehicle Technology Offices, under Joining Core Program.