Experiment and prediction model for surface roughness in micro-milling

An orthogonal experiment regression analysis and a response surface methodology are used to build the models to predict roughness of aluminum surface machined by a micro turn-milling NC machine. The influence of milling parameters used in the experiment is analyzed by the two means, orthogonal analysis and RSM. The milling parameters include cutting speed, feed per tooth, and cutting depth. In contrast with the orthogonal analysis, the RSM is an optimization prediction model and has the higher precision in micro-milling. The significance order of the parameters in the prediction model is determined based on the result of the experiment. The cutting speed has the most significant effect on surface roughness, and the second and the third significant parameters are feed per tooth and the cutting depth respectively by the rounded analysis in the current experimental condition. The RSM prediction model has higher fitting degree and practicability than the orthogonal analysis method. The milling parameters can be chosen to control and improve the quality of the surface roughness based on the prediction model of RSM.