Integrated Dynamic Optimization and Scheduling of Polymerization Processes with First Principle Models

Abstract

An integrated dynamic optimization strategy is presented for block copolymerization processes. First the reactor model, which has unstable modes and may lead to unbounded profiles under certain design and operating conditions, is derived. Second, optimal recipes for operating of the copolymerization process are determined. Finally, an optimization strategy for the integration of scheduling and dynamic process operation for general continuous/batch processes is considered. The resulting approach leads to significant improvements in productivity, while maintaining safe operation and satisfaction of product specifications.

Discrete-Time Network Scheduling and Dynamic Optimization of Batch Processes with Variable Processing Times through Discrete-Steepest Descent Optimization

“…Although it is a local algorithm, the proposed D-SDA relies on integral convexity from discrete-convex analysis rather than relying on the convexity of the relaxed MINLP to guarantee global optimality. For instance, the D-SDA is expected to result in enhanced performance when compared to the generalized Benders decomposition (GBD) strategy, which is the state-of-the-art solution strategy employed in the solution of network SSDO problems, e.g., see Chu and You , and Nie et al , To illustrate this, a simple MINLP problem is discussed in Section S4 of the Supporting Information. This illustrative MINLP example shows that when both D-SDA and GBD are initialized from the same point, the D-SDA converges to the global optimum of the problem, while GBD stagnates at the initialization.…”