Abstract: In order to be able to more accurately predict the height of fracture when the sheet metal is formed, this paper takes the Johnson-Cook damage criterion as the standard, adopts the principal component analysis method, measures the fracture data by orthogonal experiment method, and uses SPSS and MALTAB data analysis software to specifically interpret the impact of the six process parameters of tool diameter, layer spacing, forming angle, sheet thickness, forming height and part size on the fracture. Construct a predicted linear regression equation for progressively formed sheet rupture. The results show that the principal component regression model based on the measured data of 16 workpieces predicts that the fracture height of the two test pieces is −10.753 0 mm and −10.815 0 mm, respectively, and the error between the actual fracture height is only 3.84% and 6.14%. It can be seen that the plate fracture model has a certain feasibility and accuracy, which can better predict the height of sheet fracture and prevent forming defects as much as possible.