A VERY SIMPLE STRUCTURE FOR NEURAL-NETWORK CONTROL OF DISTILLATION

This paper presents a novel approach for process control that uses neu ral networks to model the steady-state inverse of a process which is t hen coupled with a simple reference system synthesis to generate a mul tivariable controller. The control strategy is applied to dynamic simu lations of two methanol-water distillation columns that express distin ctly different behaviour from each other (one simulates a lab column, while the second simulates an industrial-scale high-purity column). A steady-state process simulation package was used to generate all the n eural network training data. An efficient training algorithm based on a nonlinear least-squares technique was used to train the networks. Th e neural network model-based controllers show robust performance for b oth setpoints and disturbances, and performed better than conventional feedback proportional-integral (PI) controllers with feedforward feat ures.