Nonlinear process model based control and optimization of a model IV FCC unit

Abstract

Nonlinear process model based control (PMBC) has been applied to the Amoco/Lehigh University Model IV FCC industrial challenge problem. A dynamic simulator which was originally written in ACSL (Advanced Continuous Simulation Language) was converted into FORTRAN and benchmarked against the open-loop responses provided in the problem statement. The riser model in the simulator was replaced by a plug flow reactor model and a simple yield model was added. This addition made it possible to apply a simplified optimization analysis as well. Nonlinear model based control was applied for the control of reactor temperature, regenerator temperature and the flue gas oxygen concentration. The temperature of the feed entering the reactor riser, the catalyst circulation rate and the regenerator air flowrate were used as manipulated variables. The dynamic macroscopic oxygen and energy balances in the regenerator and the steady state energy balance for the reactor were used as nonlinear models for control. The Generic Model Control (GMC) law was used for the nonlinear PMBC controller. The nonlinear PMBC and conventional proportional-integral (PI) controllers were tested first for the unconstrained control. The PI results are included as a point of reference. The nonlinear PMBC constraint controllers were developed for constraints on the feed preheater, the air flowrate to the regenerator and the catalyst circulation rate. The feed rate to the system was manipulated based upon maximizing the process throughput without violating any of the constraints. The constraint controllers were tested successfully for a set of disturbances. The same controller models were then used for optimization studies to analyze the operation at the economic optimum in the face of variations in feed characteristics and variations in operative constraints.