Finite Horizon MPC for Systems in Innovation Form

John Bagterp Jørgensen; Jakob Kjøbsted Huusom; James B. Rawlings

doi:10.1109/CDC.2011.6161509

Finite Horizon MPC for Systems in Innovation Form

University of Wisconsin-Madison

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

System identification and model predictive control have largely developed as two separate disciplines. Nevertheless, the major part of industrial MPC commissioning is generation of data and identification of models. In this contribution we attempt to bridge this gap by contributing some of the missing links. Input-output models (FIR, ARX, ARMAX, Box-Jenkins) as well as subspace models can be represented as state space models in innovation form. These models have correlated process and measurement noise. The correct LQG control law for systems with correlated process and measurement noise is not well known. We provide the correct finite-horizon LQG controller for this system and use this to develop a state space representation of the closed-loop system. This representation is used for closed-loop frequency and covariance analysis. These measures are used in tuning of the unconstrained and constrained MPC. We demonstrate our results on a simulated industrial furnace.

Original language	English
Title of host publication	Proceeding of the 50th IEEE Conference on Decision and Control and European Control Conference : IEEE CDC-ECC 2011
Publisher	IEEE
Publication date	2011
Pages	1896-1903
ISBN (Print)	978-1-61284-799-3
DOIs	https://doi.org/10.1109/CDC.2011.6161509
Publication status	Published - 2011
Event	50th IEEE Conference on Decision and Control and European Control Conference - Hilton Orlando Bonnet Creek, Orlando, Florida, United States Duration: 12 Dec 2011 → 15 Dec 2011 Conference number: 50 http://control.disp.uniroma2.it/cdcecc2011/

Conference

Conference	50th IEEE Conference on Decision and Control and European Control Conference
Number	50
Location	Hilton Orlando Bonnet Creek
Country/Territory	United States
City	Orlando, Florida
Period	12/12/2011 → 15/12/2011
Sponsor	Society for Industrial and Applied Mathematics
Internet address	http://control.disp.uniroma2.it/cdcecc2011/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1109/CDC.2011.6161509

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{a7130f76a0194e46be76fcac5f773b87,

title = "Finite Horizon MPC for Systems in Innovation Form",

abstract = "System identification and model predictive control have largely developed as two separate disciplines. Nevertheless, the major part of industrial MPC commissioning is generation of data and identification of models. In this contribution we attempt to bridge this gap by contributing some of the missing links. Input-output models (FIR, ARX, ARMAX, Box-Jenkins) as well as subspace models can be represented as state space models in innovation form. These models have correlated process and measurement noise. The correct LQG control law for systems with correlated process and measurement noise is not well known. We provide the correct finite-horizon LQG controller for this system and use this to develop a state space representation of the closed-loop system. This representation is used for closed-loop frequency and covariance analysis. These measures are used in tuning of the unconstrained and constrained MPC. We demonstrate our results on a simulated industrial furnace.",

author = "J{\o}rgensen, \{John Bagterp\} and Huusom, \{Jakob Kj{\o}bsted\} and Rawlings, \{James B.\}",

year = "2011",

doi = "10.1109/CDC.2011.6161509",

language = "English",

isbn = "978-1-61284-799-3",

pages = "1896--1903",

booktitle = "Proceeding of the 50th IEEE Conference on Decision and Control and European Control Conference",

publisher = "IEEE",

address = "United States",

note = "50th IEEE Conference on Decision and Control and European Control Conference, CDC-ECC ; Conference date: 12-12-2011 Through 15-12-2011",

url = "http://control.disp.uniroma2.it/cdcecc2011/",

}

Jørgensen, JB , Huusom, JK & Rawlings, JB 2011, Finite Horizon MPC for Systems in Innovation Form. in Proceeding of the 50th IEEE Conference on Decision and Control and European Control Conference: IEEE CDC-ECC 2011. IEEE, pp. 1896-1903, 50th IEEE Conference on Decision and Control and European Control Conference, Orlando, Florida, United States, 12/12/2011. https://doi.org/10.1109/CDC.2011.6161509

Finite Horizon MPC for Systems in Innovation Form. / Jørgensen, John Bagterp ; Huusom, Jakob Kjøbsted; Rawlings, James B.
Proceeding of the 50th IEEE Conference on Decision and Control and European Control Conference: IEEE CDC-ECC 2011. IEEE, 2011. p. 1896-1903.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Finite Horizon MPC for Systems in Innovation Form

AU - Jørgensen, John Bagterp

AU - Huusom, Jakob Kjøbsted

AU - Rawlings, James B.

N1 - Conference code: 50

PY - 2011

Y1 - 2011

N2 - System identification and model predictive control have largely developed as two separate disciplines. Nevertheless, the major part of industrial MPC commissioning is generation of data and identification of models. In this contribution we attempt to bridge this gap by contributing some of the missing links. Input-output models (FIR, ARX, ARMAX, Box-Jenkins) as well as subspace models can be represented as state space models in innovation form. These models have correlated process and measurement noise. The correct LQG control law for systems with correlated process and measurement noise is not well known. We provide the correct finite-horizon LQG controller for this system and use this to develop a state space representation of the closed-loop system. This representation is used for closed-loop frequency and covariance analysis. These measures are used in tuning of the unconstrained and constrained MPC. We demonstrate our results on a simulated industrial furnace.

AB - System identification and model predictive control have largely developed as two separate disciplines. Nevertheless, the major part of industrial MPC commissioning is generation of data and identification of models. In this contribution we attempt to bridge this gap by contributing some of the missing links. Input-output models (FIR, ARX, ARMAX, Box-Jenkins) as well as subspace models can be represented as state space models in innovation form. These models have correlated process and measurement noise. The correct LQG control law for systems with correlated process and measurement noise is not well known. We provide the correct finite-horizon LQG controller for this system and use this to develop a state space representation of the closed-loop system. This representation is used for closed-loop frequency and covariance analysis. These measures are used in tuning of the unconstrained and constrained MPC. We demonstrate our results on a simulated industrial furnace.

U2 - 10.1109/CDC.2011.6161509

DO - 10.1109/CDC.2011.6161509

M3 - Article in proceedings

SN - 978-1-61284-799-3

SP - 1896

EP - 1903

BT - Proceeding of the 50th IEEE Conference on Decision and Control and European Control Conference

PB - IEEE

T2 - 50th IEEE Conference on Decision and Control and European Control Conference

Y2 - 12 December 2011 through 15 December 2011

ER -

Finite Horizon MPC for Systems in Innovation Form

Abstract

Conference

UN SDGs

Access to Document

OpenUrl availability

Fingerprint

Cite this