Dynamic Modeling and Advanced Control of a Refinery Hydrocracker Proces

Development of a dynamic non-isothermal hydrocracker model that can estimate temperatures and concentration profile along the reactor is undertaken in this work. With reaction kinetics being critical to modeling and simulation of a hydrocracking reactor, the complex chemistry of hydrocarbons is represented by a continuous lumping approach. The true boiling points of the mixture are used as characterization parameters.;A parameter estimation framework is developed for estimating the key kinetic parameters for the continuous kinetic lumping approach. Experimental data for Maya crude oil obtained before and after cracking through a bench-scale hydrocracking reactor are used to obtain parameters for the continuous kinetic lumping technique.;The model is used to study the impact of a number of key variables. Finally a model-based controller is developed for controlling the weight fractions of gasoline, and/or kerosene, and/or diesel in the final product. A linear model is identified and used in the model-based control. Both single input single output (SISO) and multiple input multiple output (MIMO) controllers are developed. Servo control and disturbance rejection characteristics of the controllers are studied.