2020 Virtual AIChE Annual Meeting

(351al) Separating Mixture Glycols in DWC with Flow Rate – Composition Cascade Control Structure

Authors

Wu, Y. - Presenter, East China University of Science and Technology
Xu, J., East China University of Science and Technology
Chen, K., East China University of Science and Technology
Wu, B., East China University of Science and Technology
Ji, L., East China University of Science and Technology
Dividing wall column (DWC) is an enhanced distillation technology which can separate multicomponent mixtures in one column with only a single condenser and reboiler. It can save up to 30% energy and reduce 30% capital costs over conventional distillation column sequences. However, there are only a few DWCs applied in industry since its first application in 1985 by BASF. The design of control structure is one of the keys to limit its industrial application.

In this work, glycols (1,2-propanediol+ 1,3-butanediol+ 1,4-butanediol) were taken as the feed mixture, an optimized 50-stages DWC was built in Aspen Plus, and four control structures were designed in Aspen Dynamics. In order to investigate the dynamic responses for the control structures in DWC, ± 10% feed disturbances would be applied to the system after 1h operation.

The first control structure proposed for DWC was temperature control structure (CS1). Reflux ratio (RR), side stream flow rate (S) and reboiler heat duty (QR) were selected as manipulated variables. Sensitivity analysis and singular value decomposition (SVD) criteria were used to obtain the appropriate reference stage. Temperature controllers were tuned through the relay-feedback test and Tyreus-Luyben tuning method. The dead time was set as 1min and the control loops were tuned sequentially. The results showed that CS1 had difficulty in maintaining the product purities under ± 10% feed disturbances, which included feed flow rate disturbances and feed composition disturbances.

Temperature difference control structure (CS2) was also investigated in this work. Different from CS1, the system could be stabilized in 3hrs for CS2 and the maximum deviations for the purities of 1,2-propanediol and 1,3-butanediol were 0.04% and 1.03%, respectively, while the purity of the bottom product 1,4-butanediol couldn’t become steady after 10hrs’ operation yet.

For Composition control structure (CS3), composition controllers were tuned through the relay-feedback test and Tyreus-Luyben tuning method. The dead time was set as 3mins. The results showed that the maximum deviations for the product purities of 1,2-propanediol, 1,3-butanediol, and 1,4-butanediol were 0.35%, 2.78% and 0.41%, respectively. The feed composition disturbances could be handled well in 4hrs while it took about 7hrs to become steady for the feed flow rate disturbances.

A flow rate-composition cascade control structure (CS4) was presented in this paper finally. A cascade loop was used to control the purity of side stream product by manipulating S. A temperature controller was used to control the temperature of reference stage in the rectifying section by manipulating RR. And a composition controller was used to control the purity of bottom product by manipulating QR. The maximum deviations for the product purities of 1,2-propanediol, 1,3-butanediol, and 1,4-butanediol were 0.07%, 1.44% and 0.46%, respectively. The DWC could recover to steady state in 4hrs. It showed that CS4 had a better performance on dealing with the disturbances than CS3.

In conclusion, the feed disturbances couldn’t be handled well for CS1; the settling time was shortest while the bottom product purity couldn’t become steady after 10hrs’ operation yet for CS2; the system could be stable with a long settling time for CS3; and the feed disturbances could be dealt with small deviations and short settling time for CS4. Both CS3 and CS4 could handle the disturbances, while CS4 had the best performance among the four control structures proposed in this work.