A mathematical model for prediction of monomer conversion, of particle number and of the evolution of the particle size distribution (PSD) in dispersion polymerization is developed. Despite being completed very early during the polymerization process (monomer conversion <1%), nucleation of new particles is the most important factor affecting the PSD. In order to describe the particle nucleation phenomena, the mechanism of homogeneous coagulative nucleation is considered. According to this mechanism, polymer chain aggregates can either coagulate and grow, to give birth to new polymer particles (particle nucleation), or be captured by existing polymer particles. Two sets of population balance equations are used: one for the aggregates, and a second one for the stable polymer particles. It is shown that the model is able to describe the dispersion polymerization of styrene in ethanol and the formation of micron-size monodisperse polymer particles.

The objective of this work is to develop a strategy of nonlinear model predictive control for industrial slurry reactors of propylene polymerizations. The controlled variables are the melt index (polymer quality) and the amount of unreacted monomer (productivity). The model used in the controller presents a linear dynamics and a nonlinear static gain given by a neuronal network MLP (multilayer perceptron). The simulated performance of the controller was evaluated for a typical propylene polymerization process. It is shown that the performance of the proposed control strategy is much better than the one obtained with the use of linear predictive controllers for setpoint tracking control problems.

Particle size distribution (PSD) of polystyrene particles produced by suspension polymerization is of fundamental importance in determining suspension stability and product quality attributes. Within a population balance framework, a model is proposed for suspension polymerization reactors to describe the evolution of the PSD. The model includes description of breakage and coalescence rates in terms of reaction kinetics and rheology of the dispersed phase. The model is validated with experimental data of styrene suspension polymerization.

or multivariable non linear predictive control implementations, a hybrid-neural model (lumped model) was successfully used for modeling a loop-tubular polymerization reactor (a lumped or distributed model, depending on recycle ratio). Bifurcation diagrams were computed in order to investigate the agreement between process and model, of paramount importance for model based controller implementation purposes. Performance was evaluated considering the nonlinear model predictive control of both a loop tubular reactor (lumped) SISO problem and a tubular reactor (distributed) MIMO problem.

Variable correlations are usually neglected during parameter estimation. Very frequently these are gross assumptions and may potentially lead to inadequate interpretation of final estimation results. For this reason, variable correlation and model parameters are sometimes estimated simultaneously in certain parameter estimation procedures. It is shown, however, that usually taking variable correlation into consideration during parameter estimation may be inadequate and unnecessary, unless independent experimental analysis of measurement procedures is carried out.