Multiple regression model of process for aluminum alloy self-piercing riveting based on ball-shaped die

The self-piercing riveting (SPR) technology based on ball-shaped die is proposed to solve the problem of cracks in the SPR process of aluminum alloy sheet materials with low ductility. The Box-Behnken Design (BBD) response surface experimental method was used to study the SPR joint process for 7075 aluminum alloy material, and the optimal process parameters were determined. Taking rivet length, punch speed and punch stroke as input variables and joint failure load and energy absorption as response values, a multiple nonlinear regression model for SPR process based on ball-shaped die was established and verified by experiments. The results show that the use of ball-shaped die can effectively restrain the crack in the mechanical inner locking zone of the SPR joint under the guarantee of joint strength. The relative error between the theoretical predicted value obtained by the regression model and the experimental value is low, and the regression model has high reliability. Meanwhile, single factor analysis shows that the rivet length has the greatest influence on the static mechanical property of the joint. In the interaction analysis, the interaction of rivet length and punch stroke has the most significant influence on the failure load and energy absorption value. Finally, NSGA-II genetic algorithm was used to determine the optimal process parameter combination.