Application of Independent Components Analysis in a Distillation Column High Purity for the Improvement of Variables Decoupling

Distillation columns constitute a significant fraction of the investment and production cost of chemical plants. Therefore need to be designed and controlled to keep yourself in optimal operating conditions. The work is in the application of ICA (Independent Component Analysis) in a multicomponent distillation column, high purity, where the mixture presents an unconventional azeotropic behavior. Two components in the plant require control for the best performance and safe operation of the plant, they are, 1.2 dichloroethane (1.2 EDC) in the top column and carbon tetrachloride (CCl4) in the base column. The column presents difficulties in stabilizing these components, which can cause problems in subsequent process units. The use of conventional decoupled projects in process control, most often do not have perfect results, especially in this type of column, due to the strong coupling between the variables. The use of artificial intelligence techniques such as ICA technique showed a significant improvement in decoupling of control loops that column, facilitating control of variables in their respective setpoints.