東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Probabilistic approach for MPC perfo...

Agarwal, Nikhil.

Linked to FindBook

Google Book

Amazon

博客來

Probabilistic approach for MPC performance assessment.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Probabilistic approach for MPC performance assessment./
Author:	Agarwal, Nikhil.
Description:	120 p.
Notes:	Source: Masters Abstracts International, Volume: 46-02, page: 1129.
Contained By:	Masters Abstracts International46-02.
Subject:	Systems science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MR29925
ISBN:	9780494299258

Probabilistic approach for MPC performance assessment.
Agarwal, Nikhil.

Probabilistic approach for MPC performance assessment. - 120 p.

Source: Masters Abstracts International, Volume: 46-02, page: 1129.

Thesis (M.Sc.)--University of Alberta (Canada), 2007.

The performance of the MPC controllers can be improved by increasing the degrees of freedom (DoF) for control purposes. The degrees of freedom can be increased by relaxing the constraints or by reducing the variance of the process variables. The Linear-Quadratic (LQ) optimization method can be used for providing guidelines for increasing the DoFs for a controller. As the LQ optimization considers mean operating point and the processes do not always operate on but around the mean operating point, it is essential to take into account the variability or distribution. Due to the presence of variability the process variables have the probabilities to be inside and outside the constraint limits. In this thesis, the Bayesian method is utilized to take into account these probabilities for the assessment of the decisions related to increasing the controller DoFs. The algorithm, for Bayesian analysis, discussed in this thesis can also be used to obtain the guidelines for increasing the controller DOFs to achieve certain level of performance. By extending this idea a probabilistic optimization technique is also introduced in this thesis. The optimization function defined in probabilistic optimizer (PO) takes into consideration the probabilities for the data distribution and the profit/loss terms associated with the distribution. The PO can be used to obtain the constraint tuning guidelines for the controllers. Extending the idea of applying Bayesian methods for LQ optimization, a tool is developed for assessing the decisions, based on PO approach, for increasing controller DoF.

ISBN: 9780494299258Subjects--Topical Terms:

3168411
Systems science.

Probabilistic approach for MPC performance assessment.
LDR:02371nmm a2200253 4500 001 2064166
005 20151109121441.5
008 170521s2007 ||||||||||||||||| ||eng d
020 $a 9780494299258
035 $a (MiAaPQ)AAIMR29925
035 $a AAIMR29925
040 $a MiAaPQ $c MiAaPQ
100 1 $a Agarwal, Nikhil. $3 3178726
245 1 0 $a Probabilistic approach for MPC performance assessment.
300 $a 120 p.
500 $a Source: Masters Abstracts International, Volume: 46-02, page: 1129.
502 $a Thesis (M.Sc.)--University of Alberta (Canada), 2007.
520 $a The performance of the MPC controllers can be improved by increasing the degrees of freedom (DoF) for control purposes. The degrees of freedom can be increased by relaxing the constraints or by reducing the variance of the process variables. The Linear-Quadratic (LQ) optimization method can be used for providing guidelines for increasing the DoFs for a controller. As the LQ optimization considers mean operating point and the processes do not always operate on but around the mean operating point, it is essential to take into account the variability or distribution. Due to the presence of variability the process variables have the probabilities to be inside and outside the constraint limits. In this thesis, the Bayesian method is utilized to take into account these probabilities for the assessment of the decisions related to increasing the controller DoFs. The algorithm, for Bayesian analysis, discussed in this thesis can also be used to obtain the guidelines for increasing the controller DOFs to achieve certain level of performance. By extending this idea a probabilistic optimization technique is also introduced in this thesis. The optimization function defined in probabilistic optimizer (PO) takes into consideration the probabilities for the data distribution and the profit/loss terms associated with the distribution. The PO can be used to obtain the constraint tuning guidelines for the controllers. Extending the idea of applying Bayesian methods for LQ optimization, a tool is developed for assessing the decisions, based on PO approach, for increasing controller DoF.
590 $a School code: 0351.
650 4 $a Systems science. $3 3168411
690 $a 0790
710 2 $a University of Alberta (Canada). $3 626651
773 0 $t Masters Abstracts International $g 46-02.
790 $a 0351
791 $a M.Sc.
792 $a 2007
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MR29925