Publication detail

CARIMA versus CARMA Model in Predictive Control

Authors: Honc Daniel

Year: 2006

Type of publication: článek v odborném periodiku

Name of source: Manufacturing Engineering / Výrobné inžinierstvo

Publisher name: Technická univerzita v Košiciach

Place: Košice

Page from-to: 53-55

Titles:

Language	Name	Abstract	Keywords
cze	CARIMA versus CARMA model v prediktivním řízení	Čtenář je seznámen s konceptem prediktivního řízení. Řešení s Controller Auto-Regressive Integrated Moving-Average (CARIMA) modelem je přímočaré a regulátor má integrační character (nezanechává trvalou regulační odchylku). Tento regulátor je ovšem velmi citlivý na šum měření. Citlivost na šum je nižší, použijeme-li Auto-Regressive Moving-Average (CARMA) model. Teoretické závěry jsou demonstrovány na jednoduchém příkladu.	prediktivní řízení, CARIMA model, CARMA model, čitlivost na šum měření
eng	CARIMA versus CARMA Model in Predictive Control	Briefly theory is given to introduce readers the Model Predictive Control concept. When Controller Auto-Regressive Integrated Moving-Average (CARIMA) model is used the solution is straightforward and resulting controller has integrating character (offset-free steady state control is archived). Admittedly the controller is rather sensitive to the measurement noise. This effect is lower when Auto-Regressive Moving-Average (CARMA) model is applied. Modifications of penalization matrix and plant output prediction computation are necessary in order to be the cost function identical as for the CARIMA model (to penalize the control increments). Simple comparative example is given to demonstrate measurement noise sensitivity of predictive controllers based on both discussed models.	Model Predictive Control, CARIMA model, CARMA model, measurement noise sensitivity

Search

Login for students

Login for employees

Publication detail

Department

Map

Services

Popular Links