On-line implementation of a model predictive controller on amultivariable chemical process

摘要

An investigation into neural network model predictive control isdescribed in this paper. The control strategy developed is applied to alaboratory process to control temperature, pH and dissolved oxygen. Themain difficulties in control of this process are nonlinearity, couplingeffects among variables and long time-delay in the heat exchanger.Parallel neural models are developed from real process data for the usewith online model predictive control and off-line simulations. Theonline control results are demonstrated