|
ADAPTIVE PREDICTIVE CONTROL OF A NEUTRALIZATION PLANT USING LOCAL MODEL NETWORKS
U. Halldorsson, A. Ali, H. Unbehauen and Chr. Schmid
Control Engineering Laboratory
Dep. of Electrical Engineering and Information Sciences,
Ruhr-Universität Bochum, D-44780 Bochum, Germany
Fax: ++49 234 32 14101
E-mail: halldorsson@esr.ruhr-uni-bochum.de
This contribution investigates the application of a nonlinear model based adaptive predictive control algorithm to a pH neutralization plant. A local linear model network is responsible for the online identification of the plant. The advantage of such a model is that the information needed by the predictive control algorithm to build gradients is directly available. This aspect reduces the computational cost significantly. Results of real-time control of a laboratory-scale plant are presented.
Keywords: Nonlinear Predictive Control, Adaptive Control, pH Control, Local Linear Model Networks, Nonlinear Systems.
Session slot T-We-A03: Adaptive Control in Industry/Area code 3b : Adaptive Control and Tuning
 | 