Dynamic Optimisation of Batch Processes by Integrated Two-Time-Scale Scheme
Institute of Information Engineering, Automation, and Mathematics, Faculty of Chemical and Food Technology, Slovak University of Technology, Radlinského 9, 812 37 Bratislava
a Department of Chemical Engineering, McMaster University, Canada
E-mail: * email@example.com
Abstract: The theory of neighbouring-extremal control has been developed over the last 4-5 decades to avoid the costly reoptimisation of dynamic systems, primarily in applications with fast non-linear dynamics. Perhaps the biggest drawback with this approach, when applied to chemical processes, is its poor performance in the presence of large parametric and structural model mismatch. On the other hand, model predictive control (MPC) and run-to-run optimisation are more resistant to model mismatch, but require time-consuming on-line reoptimisation that restricts their applications to slow dynamic systems. This paper proposes to combine both approaches in order to mitigate their deficiencies, thereby leading to an integrated two-time-scale scheme with enhanced performance and tractability for dynamic real-time optimization. This scheme is demonstrated by two batch reactor examples.
Keywords: dynamic optimisation, neighbouring extremals, optimal control, two-time-scale scheme
Full paper in Portable Document Format: acs_0067.pdf
Acta Chimica Slovaca, Vol. 3, No. 2, 2010, pp. 38—56