(226d) Optimal Temperature Control Of An Industrial Scale Batch Chemical Reactor With Regard To Swelling

The goal of this work is the optimization of the temperature profile in a large scale industrial reactor with the swelling constraint using optimal control techniques. The reaction system is a large scale, complex three-phase (solid-liquid-vapor) mixture with equilibrium reactions. The reaction leads to the production of the main product (P) and to a gas co-product (CP). The operation temperature is constrained by a maximal formation rate of the CP. When the rate of formation is too high the produced gas entrains the reaction mass which swells over the reactor level and can reach in the pipes. This phenomenon is highly undesired due to the fact that leads to reactor shut down and pipe cleaning, thus productivity loss. In this work the aim is to calculate the safest temperature profile, in this way compensating the modeling errors. From chemical modeling point of view this is achieved by formulating a reaction kinetic model that is based on four forward kinetic equations. This kinetic model was identified based on concentration measurements and it is considered that has good accuracy. The kinetic model is linked with the bubbly vessel hydrodynamic model which predicts the swelling height in function of reaction rate, temperature, pressure and physical property of the reaction mixture. The chosen hydrodynamic model is the most conservative, which means that it will predict the highest level rise in the reactor. The hydrodynamic model was tested using CP gas rates that were formed during regular production process. The level predicted by the hydrodynamic model showed a good concordance with the observed reactor level therefore it is considered adequate for the modeling and optimization purposes. The maximum reaction rate was calculated using optimal control optimization, with the temperature as control variable. The objective was to maximize productivity of the desired product without to exceed the maximum reactor level. The non-linear optimization constraints were formulated in the form of the dynamic reaction model and the hydrodynamic model. The modeling and optimization results show that the currently implemented profile is conservative and a reactor performance improvement of up to 36% can be achieved by implementing the optimal temperature profile.