Nonlinear Model-based Multivariable Control of Refinery Processes

In order to demonstrate the effectiveness of nonlinear model-based control with regards to industrial application, a nonlinear control strategy was developed for an industrial debutanizer and implemented in real-time. The nonlinear control provided improved control performance of the product qualities. A steady-state model of debutanizer with approximate dynamics was used in the control strategies. A complex multivariable control problem was solved by formulating the nonlinear constrained optimization strategy for a crude distillation process. The heavy oil fractionator problem proposed by Shell Oil was selected for this work. The method uses the nonlinear model-based controller, which considers the model uncertainty explicitly.