Analysis And Model Prediction Of The Surface Roughness For Hardened Steel Die Corner In Annular Milling Process

Automobile cover parts die is a sign of automobile manufacturing level with main difficulties of high hardness, complex configuration and high surface accuracy. The cavity corner is a typical structure of automobile cover parts die, its surface roughness is an important indicator for measuring the quality of die cavity surface.In this paper, the annular milling cutter is used for milling the cavity corners. Compared with the point contact and narrow processing characteristics of ball milling cutter, the annular milling cutter has a larger contact area and higher rigidity, which could reduce working hours and achieve a higher surface machining accuracy by improving milling efficiency. In reponse to the low surface quality in die cavity corner machining, we studied the influence relations between the surface roughness of annular milling corner and milling technological parameters on a plane and a 30°inclined plane were studied by the orthogonal experiments, respectively; then established two prediction models on annular milling corner surface roughness and their predictive effects were compared. Based on the analysis of influencing factors for annular milling corner's surface quality, the validation experiment was carried out by using optimized milling parameters, which tested the accuracy of prediction model and the aim of improving die surface quality.Taking chilled steel Cr12 Mo V as the research object and aiming at main milling parameters, including spindle speed, feed speed, back engagement and corner angle, the L16(44) orthogonal experiments for annular milling corner on a plane and a 30°inclined plane were designed, and the surface roughness of corner in different areas was measured and characterized by a laser confocal microscopy.By analyzing each surface roughness data in the orthogonal experiment, the effect of milling technological parameters on corner surface roughness and the main influence factors was carried out. Meanwhile, the optimized technology combinations of annular milling corner on a plane and a 30°inclined plane were determined.Based on the Matlab2013, the parameters of multivariate linear regression and least squares support vector machines(LS-SVM) prediction model were calculated, establishing the multivariate linear regression and LS-SVM prediction models of surface roughness in annular milling corner. Using the measured data and the predicted proximity, introducing of error evaluation factors.Then further analysis and comparison on the predictive effect and error between two models were carried out.In order to verify the accuracy LS-SVM surface roughness prediction model, the annular milling process adopting optimized technology parameters on a plane and a 30°inclined plane was conducted. Through the milling force and surface roughness test of annular milling 120 degrees corner, it can be found that optimized technology parameters could reduce the dramatic change of milling force in corner area. The measured values of surface roughness also can verify the accuracy and effectiveness of the roughness prediction model.