KAMAL IQBAL, NUR HIDAYAH
(2015)
MODEL-PLANT MISMATCH DETECTION, ISOLATION AND
ESTIMATION USING MISMATCH MODELING.
Masters thesis, Universiti Teknologi PETRONAS.
Abstract
The performance of MPC highly depends on the accuracy of the model of the
plant used in the design of MPC. The difference between the model of the plant and
the actual dynamic and steady state characteristics of the plant is called model plant
mismatch (MPM). Even though MPC can tolerate some amount of mismatch,
excessive MPM can significantly deteriorate the performance of the controller. In
MIMO systems
re-identification of complete plant is very costly and time consuming. This research
aims at developing a method for detecting submodels that have significant mismatch,
estimating the magnitude of the mismatch and isolating those submodels that have
significant mismatches to deteriorate the performance of the controller.
Actions (login required)
 |
View Item |