D-ADMM based distributed MPC with input-output models

Rafael P. Costa; João M. Lemos; João F. C. Mota; João M. F. Xavier

doi:10.1109/CCA.2014.6981422

D-ADMM based distributed MPC with input-output models

Rafael P. Costa, João M. Lemos, João F. C. Mota, João M. F. Xavier

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

9 Citations (Scopus)

Abstract

This article presents a distributed model predictive controller (MPC) based on linear models that use input/output plant data and D-ADMM optimization. The use of input/output models has the advantage of not requiring a Kalman filter to estimate the plant state. The D-ADMM algorithm solves the optimization problem associated to a cost function that is the sum of the control agents private costs, being a modification of the Alternating Direction of Multipliers (ADMM) algorithm that requires no central node and implies a significant reduction in the communication among adjacent nodes. The distributed MPC is obtained for the special case of a linear graph. An application to distributed control of a water delivery canal is presented to illustrate the algorithm.

Original language	English
Title of host publication	2014 IEEE Conference on Control Applications, CCA 2014
Publisher	IEEE
Pages	699-704
Number of pages	6
ISBN (Electronic)	9781479974092
DOIs	https://doi.org/10.1109/CCA.2014.6981422
Publication status	Published - Dec 2014
Event	2014 IEEE Conference on Control Applications - Juan Les Antibes, France Duration: 8 Oct 2014 → 10 Oct 2014

Conference

Conference	2014 IEEE Conference on Control Applications
Abbreviated title	CCA 2014
Country/Territory	France
City	Juan Les Antibes
Period	8/10/14 → 10/10/14

ASJC Scopus subject areas

Computer Science Applications
Control and Systems Engineering

Access to Document

10.1109/CCA.2014.6981422

Cite this

@inproceedings{e369ed405f0b4f048eee25d2f66bf5f1,

title = "D-ADMM based distributed MPC with input-output models",

abstract = "This article presents a distributed model predictive controller (MPC) based on linear models that use input/output plant data and D-ADMM optimization. The use of input/output models has the advantage of not requiring a Kalman filter to estimate the plant state. The D-ADMM algorithm solves the optimization problem associated to a cost function that is the sum of the control agents private costs, being a modification of the Alternating Direction of Multipliers (ADMM) algorithm that requires no central node and implies a significant reduction in the communication among adjacent nodes. The distributed MPC is obtained for the special case of a linear graph. An application to distributed control of a water delivery canal is presented to illustrate the algorithm.",

author = "Costa, \{Rafael P.\} and Lemos, \{Jo{\~a}o M.\} and Mota, \{Jo{\~a}o F. C.\} and Xavier, \{Jo{\~a}o M. F.\}",

year = "2014",

month = dec,

doi = "10.1109/CCA.2014.6981422",

language = "English",

pages = "699--704",

booktitle = "2014 IEEE Conference on Control Applications, CCA 2014",

publisher = "IEEE",

address = "United States",

note = "2014 IEEE Conference on Control Applications, CCA 2014 ; Conference date: 08-10-2014 Through 10-10-2014",

}

TY - GEN

T1 - D-ADMM based distributed MPC with input-output models

AU - Costa, Rafael P.

AU - Lemos, João M.

AU - Mota, João F. C.

AU - Xavier, João M. F.

PY - 2014/12

Y1 - 2014/12

N2 - This article presents a distributed model predictive controller (MPC) based on linear models that use input/output plant data and D-ADMM optimization. The use of input/output models has the advantage of not requiring a Kalman filter to estimate the plant state. The D-ADMM algorithm solves the optimization problem associated to a cost function that is the sum of the control agents private costs, being a modification of the Alternating Direction of Multipliers (ADMM) algorithm that requires no central node and implies a significant reduction in the communication among adjacent nodes. The distributed MPC is obtained for the special case of a linear graph. An application to distributed control of a water delivery canal is presented to illustrate the algorithm.

AB - This article presents a distributed model predictive controller (MPC) based on linear models that use input/output plant data and D-ADMM optimization. The use of input/output models has the advantage of not requiring a Kalman filter to estimate the plant state. The D-ADMM algorithm solves the optimization problem associated to a cost function that is the sum of the control agents private costs, being a modification of the Alternating Direction of Multipliers (ADMM) algorithm that requires no central node and implies a significant reduction in the communication among adjacent nodes. The distributed MPC is obtained for the special case of a linear graph. An application to distributed control of a water delivery canal is presented to illustrate the algorithm.

UR - https://www.scopus.com/pages/publications/84920525701

U2 - 10.1109/CCA.2014.6981422

DO - 10.1109/CCA.2014.6981422

M3 - Conference contribution

AN - SCOPUS:84920525701

SP - 699

EP - 704

BT - 2014 IEEE Conference on Control Applications, CCA 2014

PB - IEEE

T2 - 2014 IEEE Conference on Control Applications

Y2 - 8 October 2014 through 10 October 2014

ER -

D-ADMM based distributed MPC with input-output models

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this