A Modeling Method to Size Concentration in Slashing Process using Principal Component Analysis

摘要

A statistical modeling method to size concentration is presented based on multi-sensor information fusion and principal component regression analysis. The criteria for selecting principal components are adopted in modeling procedure in which the relation between the signs of regression coefficients and the signs of correlation coefficients is considered as rules for principal component selection. The results of example demonstrate that a size concentration model with the lower relative errors can be obtained by using strict rules for principal component selection.