Application of optimal adaptive generalized predictive control to a packed distillation column

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Vol. 84; no. 3; pp. 389 - 396
• Main Authors: Karacan, S, Hapoǧlu, H, Alpbaz, M
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 15-12-2001; Elsevier
• Subjects: Applied sciences; Chemical engineering; Distillation; Exact sciences and technology; Generalized predictive control; Optimization; Packed distillation; ISE, integral square of error; OPT, optimum values; Packed distillation; Generalized predictive control; IAE, integral absolute of error; ARIMAX, autoreggressive integral moving average with external input; GPC, generalized predictive control; GMV, generalized minimum variance; PRBS, pseudo random binary sequence; PID, proportional-integral-derivative; Optimization; Water; Parameter estimation; Adaptive control; Modeling; Finite element method; Static model; Dynamic model; Mathematical model; Collocation method; Predictive control; Methanol; Selftuning control; Distillation; Process control; Theoretical study; Experimental study; Orthogonal transformation; Primary alcohol; Binary mixture; Experimental design; Optimal control; Numerical simulation; Temperature control; Packed column; Differential integral proportional control
• ISSN: 1385-8947; 1873-3212
• DOI: 10.1016/S1385-8947(01)00130-9

In this work, optimal operating conditions for a packed distillation column and optimal adaptive generalized predictive control (OA-GPC) were investigated. Thus, the dynamic and steady-state properties of the packed distillation column distilling methanol–water mixture were observed experimentally and theoretically. Mathematical models for the packed distillation column were solved with orthogonal collocation on finite elements. Optimal operating conditions of the system were found by using Box–Wilson optimization method and “Experimental Design” technique.
Two types of control algorithm were utilized for controlling the packed distillation column, viz. conventional proportional integral derivative (PID) and generalized predictive control (GPC) at optimal operating conditions. Overhead temperature control was examined experimentally and theoretically. Pseudo random binary sequence (PRBS) signal and recursive identification algorithm were used to estimate the relevant parameters of the polynomial ARIMAX model. Generally theoretical and experimental control results were in accord with each other and it was observed that OA-GPC produced better performance than PID for the packed distillation column.