Intensification of Emulsion Copolymerization – Development and Application of Mathematical Model for Nonlinear Model Predictive Control

摘要

Emulsion polymerization represents well-known and mature technology, in which products are often manufactured in semi-batch reactors according to a fixed recipe schedule. Although this tight and robust recipe is finalized only after a period of product and process development, it does not allow for batch-to-batch changes in order to further increase the process efficiency. However, such intensification can be achieved by employing nonlinear model predictive control (NMPC), which allows time-varying feeds and reactor temperature to run the reaction faster, operate the reactor closer to its limits, and still keep the designed product specifications (such as molecular weight distribution, copolymer composition, viscosity, etc.). In the case of emulsion polymerization, the most critical constraint is usually the cooling capacity, which significantly varies from batch to batch due to the fouling of the reactor walls.