A surface roughness prediction model of robot grinding based on orthogonal experiments

Abstract Design a three-factor and three-level orthogonal experiment to analyze the effects of belt grit, robot speed, and sand belt wheel speed on surface roughness during robot grinding a cast iron pipe tee. Analysis results show that belt grit has a significant impact on the surface roughness. We used segmented linear surfaces, segmented cubic surfaces, and biharmonic surfaces to analyze the influence of process parameters on surface roughness. The results show that the fitting effect of biharmonic surfaces was good. Finally, a surface roughness prediction model for robot grinding of cast iron parts was established based on multiple linear regression analysis.