Iterative Nonlinear Model Predictive Control of a pH Reactor. A Comparative Analysis
Abstract
This paper describes the control of a batch pH reactor by anonlinear predictive controller that improves performance by usingdata of past batches. The control strategy combines the feedbackfeatures of a nonlinear predictive controller with the learningcapabilities of run-to-run control.The inclusion of real-time data collected during the on-goingbatch run in addition to those from the past runs make the controlstrategy capable not only of eliminating repeated errors but alsoof responding to new disturbances that occur during the run. Thepaper uses these ideas to devise an integrated controller thatincreases the capabilities of NMPC with batch-wise learning. A comparison with other ontrollers is presented.
