Abstract: Establishing an effective and reliable prediction model of self-piercing riveting process and mechanical properties is an important problem to be solved in the process of industrial application and promotion. Box-Behnken Design response face test was carried out to investigate the parameter of AA5182, AA5052 and AL1420 aluminum alloy self-piercing riveting. Taking the plate thickness, material hardness and rivet hardness were used as input values. Meanwhile, the punch stroke, the maximum riveting force and the failure load of the joint were used as the output response values. The experimental was established the multiple nonlinear regression model between the influencing factors and the response value, and the influence rule of multiple input parameters for response values was studied. The result shows that, the errors between the process and strength theory prediction values of the model and the real values are within 8%. The optimized regression model has high reliability in engineering applications. Combined with the three-dimensional response surface and contour map of the model, it can be seen that the interaction of plate thickness and rivet hardness has the greatest impact on the failure load of the joint and maximum riveting force, and the punch stroke is mainly affected by the interaction of plate hardness and rivet hardness.

Key words: aluminum alloy, self-piercing riveting(SPR), regression prediction model, input parameter