A Surface Roughness Prediction Model for SKT4 Steel Milling

Abstract

Predicting surface roughness is critical in manufacturing processes like grinding, particularly for materials such as SKT4 steel, where a precise surface finish is imperative. Precise roughness prediction facilitates the optimization of process parameters to achieve the desired surface quality, consequently diminishing the need for supplementary operations such as grinding or polishing. This, in turn, decreases costs and lead times. This study aimed to develop a surface roughness prediction model tailored for milling SKT4 steel by designing experiments to analyze the influence of cutting parameters on surface roughness, collecting and analyzing data related to machining parameters in process modeling, and developing and validating the model. The analysis of variance (ANOVA) results highlights the significant influence of the interaction between rotational speed and cutting depth on skin roughness. The linear regression model demonstrates clear variability in the data (R2 of approximately 99.74%) and exhibits effective predictive capabilities (pred. R2 of approximately 83.56%). The maximum impact on skin roughness was observed at a rotational speed of 1500 RPM, a feed rate of 300 mm per minute, and a cutting depth of 0.2 mm. Increasing rotational speed leads to smoother skin, whereas higher feed rates result in decreased smoothness. However, skin roughness shows minimal fluctuation with changes in feed rate and cutting depth. The model accurately predicts the average skin roughness values.