PROJECT TITLE :

Multivariate Control for Three Variables of an Industrial Roll-Type Electrostatic Separator

ABSTRACT:

Hotelling's $T^2$ charts have already proven their effectiveness in the multivariate management of electrostatic separation method. The aim of this work is to increase the utilization of these charts from two to 3 variables, so as to enhance the statistical control of an industrial roll-type electrostatic separator designed for the recycling industry. In the case of electrostatic separation processes, the utilization of Hotelling's $T^2$ chart for 3 variables would permit controlling on a single graph the correlation between the lots accumulated within the three boxes of the merchandise collector. The results of the study administrated on a pilot-plant electrostatic separator for metals and insulators from electric cable wastes in Algeria demonstrated the effectiveness of such a chart in detecting out-of-management situations: accumulation of dirt particles on the high-voltage electrodes and reduction of the input flow of raw materials.