15th Triennial World Congress of the International Federation of Automatic Control
  Barcelona, 21–26 July 2002 
AN ON-LINE COLOR MANAGEMENT SYSTEM BASED ON A NON-LINEAR PREDICTIVE APPROACH
Daniel Sbarbaro.*
* Department of Electrical Engineering, Universidad de.
Concepción, Concepción, CHILE.
Email: dsbrbar@die.udec.cl

Color is an important variable in many processes since it defines the final appearance of the product. The color management system must not only consider the non-linearities of the process but also the operational constraints. This work proposes a Predictive Controller Approach, based on a sequential linear programming algorithm, for an on-line color recipe formulation and correction. Experiments with three colorants are presented, demonstrating the performance of the proposed algorithm.
Keywords: Predictive Controller, Coloring Process, Kubelka-Munk theory
Session slot T-Th-E11: Nonlinear Model Predictive Control/Area code 7a : Chemical Process Control