Since the fact that the surface topography of micro-milling is complex and difficult to evaluate accurately

a surface roughnessdetection index based on three-dimensional (3D) characterization is proposed. Firstly

on the basis of principal component analysis

the 3D surface roughness parameters Sa

Ssk and Sku are converted into gray correlation degree as a surface roughness detection index based on the gray correlation analysis method. Secondly

the response surface methodology (RSM) is used to establish a gray correlation degree model to analyze the influence of machining parameters on the GRG. Finally

the combination of optimal machining parameters is obtained and verified. The results show that the average relative error of the gray correlation degree model is 6.54%

the fitting accuracy is high and the prediction effect is good

which verifies the feasibility of the model. The GRG corresponding to the obtained optimal process parameter combination is increased by 15.27%

which realizes the purpose of surface roughness minimization and surface abnormal features minimization and proves the feasibility of the detection index.