Closed-loop Aspects of Data-Enabled Predictive Control

Rogier Dinkla; Sebastiaan P. Mulders; Jan Willem van Wingerden; Tom Oomen

doi:10.1016/j.ifacol.2023.10.1806

Closed-loop Aspects of Data-Enabled Predictive Control

Rogier Dinkla^*, Sebastiaan P. Mulders^*, Jan Willem van Wingerden^*, Tom Oomen^*

^*Corresponding author for this work

Research output: Contribution to journal › Conference article › Scientific › peer-review

2 Citations (Scopus)

20 Downloads (Pure)

Abstract

In recent years, the amount of data available from systems has drastically increased, motivating the use of direct data-driven control techniques that avoid the need of parametric modeling. The aim of this paper is to analyze closed-loop aspects of these approaches in the presence of noise. To analyze this, a unified formulation of several approaches, including Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) is obtained and the influence of noise on closed-loop predictors is analyzed. The analysis reveals potential closed-loop correlation problems, which are closely related to well-known results in closed-loop system identification, and consequent control issues. A case study reveals the hazards of noise in data-driven control.

Original language	English
Pages (from-to)	1388-1393
Number of pages	6
Journal	IFAC-PapersOnLine
Volume	56
Issue number	2
DOIs	https://doi.org/10.1016/j.ifacol.2023.10.1806
Publication status	Published - 2023
Event	22nd IFAC World Congress - Yokohama, Japan Duration: 9 Jul 2023 → 14 Jul 2023

Keywords

closed loop identification
Data-driven control
data-enabled predictive control
instrumental variables
subspace predictive control

Access to Document

10.1016/j.ifacol.2023.10.1806

1-s2.0-S2405896323022152-mainFinal published version, 1.03 MBLicence: CC BY-NC-ND

Cite this

@article{34130b47a2714a3391891a25cc12536e,

title = "Closed-loop Aspects of Data-Enabled Predictive Control",

abstract = "In recent years, the amount of data available from systems has drastically increased, motivating the use of direct data-driven control techniques that avoid the need of parametric modeling. The aim of this paper is to analyze closed-loop aspects of these approaches in the presence of noise. To analyze this, a unified formulation of several approaches, including Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) is obtained and the influence of noise on closed-loop predictors is analyzed. The analysis reveals potential closed-loop correlation problems, which are closely related to well-known results in closed-loop system identification, and consequent control issues. A case study reveals the hazards of noise in data-driven control.",

keywords = "closed loop identification, Data-driven control, data-enabled predictive control, instrumental variables, subspace predictive control",

author = "Rogier Dinkla and Mulders, {Sebastiaan P.} and {van Wingerden}, {Jan Willem} and Tom Oomen",

year = "2023",

doi = "10.1016/j.ifacol.2023.10.1806",

language = "English",

volume = "56",

pages = "1388--1393",

journal = "IFAC-PapersOnLine",

issn = "2405-8963",

publisher = "Elsevier",

number = "2",

note = "22nd IFAC World Congress ; Conference date: 09-07-2023 Through 14-07-2023",

}

TY - JOUR

T1 - Closed-loop Aspects of Data-Enabled Predictive Control

AU - Dinkla, Rogier

AU - Mulders, Sebastiaan P.

AU - van Wingerden, Jan Willem

AU - Oomen, Tom

PY - 2023

Y1 - 2023

N2 - In recent years, the amount of data available from systems has drastically increased, motivating the use of direct data-driven control techniques that avoid the need of parametric modeling. The aim of this paper is to analyze closed-loop aspects of these approaches in the presence of noise. To analyze this, a unified formulation of several approaches, including Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) is obtained and the influence of noise on closed-loop predictors is analyzed. The analysis reveals potential closed-loop correlation problems, which are closely related to well-known results in closed-loop system identification, and consequent control issues. A case study reveals the hazards of noise in data-driven control.

AB - In recent years, the amount of data available from systems has drastically increased, motivating the use of direct data-driven control techniques that avoid the need of parametric modeling. The aim of this paper is to analyze closed-loop aspects of these approaches in the presence of noise. To analyze this, a unified formulation of several approaches, including Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) is obtained and the influence of noise on closed-loop predictors is analyzed. The analysis reveals potential closed-loop correlation problems, which are closely related to well-known results in closed-loop system identification, and consequent control issues. A case study reveals the hazards of noise in data-driven control.

KW - closed loop identification

KW - Data-driven control

KW - data-enabled predictive control

KW - instrumental variables

KW - subspace predictive control

UR - http://www.scopus.com/inward/record.url?scp=85173828894&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2023.10.1806

DO - 10.1016/j.ifacol.2023.10.1806

M3 - Conference article

AN - SCOPUS:85173828894

SN - 2405-8963

VL - 56

SP - 1388

EP - 1393

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

IS - 2

T2 - 22nd IFAC World Congress

Y2 - 9 July 2023 through 14 July 2023

ER -

Closed-loop Aspects of Data-Enabled Predictive Control

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this