Abstract:Surface roughness of a part has a significant impact on its performance.In order to solve the problem that the single roughness parameter of the part surface cannot fully reflect the surface defects of the part,a method of applying multivariate statistical process control based on correlation analysis and principal component analysis was proposed to evalute the surface uniformity of electrochemical finishing parts.Two samples without and with surface defects were designed for comparative experiments,and a set of surface roughness parameters of the two samples were recorded after electrochemical finishing.Univariate statistical process control of multiple parameters was carried out,and then correlation analysis and principal component analysis of multiple roughness parameters were carried out to obtain the comprehensive evaluation value and to draw statistical control charts.The comparative analysis results show that for the surface with local defects,compared with the single roughness parameter,the comprehensive evaluation control chart is more accurate in detecting the surface defects of the parts.Multivariate statistical process control based on correlation analysis and principal component analysis is an effective method to evaluate the surface uniformity of electrochemical finishing parts.