Model Based Observation and Control of Distillation Columns

Abstract

Two applications of mathematical process modeling to industrial scale distillation columns are discussed. In the first part the stripping section of a methanol-water separation column is described with the help of the location of a temperature front. A simple linear state space model is used to design an observer and state controller for the reboiler duty of the column. Its application is compared with experimental data obtained from an industrial distillation column with a PID-controller.In the second part an analytically reduced mathematical model for the estimation of the product composition in multicomponent distillation is presented. It is applied successfully to a fractionation plant consisting of three interlinked columns and is used for the purpose of diagnosis.