Robust modeling and optimization of borehole enlarging by helical milling of aluminum alloy Al7075

Abstract

The paper aims to analyze the use of helical milling in hole enlargement in aluminum alloys, considering the effects of the input variables (tool overhang, cutting speed, axial and tangential feed per tooth) on the output ones related to the hole quality such as radial force, circularity and roughness. Additionally, the robust optimization is developed through the combination of robust parameter design, mean square error, and normal boundary intersection through responses modeled by response surface methodology. The results show that hole enlargement by helical milling is feasible and could achieve good values of radial forces, circularities, and specially roughness, which presented average roughness values under 0.2 μm for all experiments. The observed relationship between input and outputs shows that the axial and tangential feed are significant for all responses in linear, quadratic, and interaction terms, and cutting speed and tool overhang have at least significant terms in linear, quadratic, or interaction models. Low values of axial feed per tooth reduce radial force and circularity. Tool overhang is a noise variable and is directly connected to the shape error, since high values increase circularity deviations. Since significant noise terms exist, the robust optimization was able to reduce the variability of tool overhang in the responses and achieve mean values of circularity, radial forces, average and total roughness of about 7 μm, 20 N, 0.1 μm, and 1.7 μm, respectively.