LONG-RANGE PREDICTIVE CONTROL OF AN ABSORPTION PACKED-COLUMN

This paper is concerned with modelling and long-range predictive contr ol of an absorption packed column. The absorption column is used to de crease the concentration of CO2 in a gas mixture below a desired value . A solution of Diethanolamine (DEA) is used as the absorbent. The pow rate of the absorbent and the concentration of CO2 in the gas mixture are selected, respectively, as control and controlled variables. A ph ysically based model has been developed from considerations of mass ba lances, of transport phenomena, and of chemical reaction in the liquid phase. It consists of three nonlinear partial differential equations. The control design is based on a simple low-order linear discrete mod el with unknown and possibly time-varying parameters. The parameters o f this model are estimated using the least squares algorithm taking in to account the requirements for long-tern adaptive control An extended horizon control policy,, based on the minimization of a quadratic cri terion function of the input and output tracking errors, is used for f eedback control. Several simulation studies highlight the applicabilit y of the involved adaptive control algorithm.