Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Autoren

  • Jörg Bremer
  • Johannes Gerster
  • Birk Brückner
  • Marcel Sarstedt
  • Sebastian Lehnhoff

Externe Organisationen

  • Carl von Ossietzky Universität Oldenburg
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Titel des SammelwerksHighlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection
UntertitelInternational Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings
Herausgeber/-innenFernando De La Prieta, Alia El Bolock, Dalila Durães, João Carneiro, Fernando Lopes, Vicente Julian
ErscheinungsortCham
Seiten191-202
Seitenumfang12
ISBN (elektronisch)978-3-030-85710-3
PublikationsstatusVeröffentlicht - 27 Sept. 2021
Veranstaltung19th International Conference on Practical Applications of Agents and Multi-Agent Systems -
Dauer: 6 Okt. 20218 Okt. 2021
https://www.paams.net/

Publikationsreihe

NameCommunications in Computer and Information Science
Band1472 CCIS
ISSN (Print)1865-0929
ISSN (elektronisch)1865-0937

Abstract

In future energy systems, decentralized control will require delegation of liabilities to small energy resources. Distributed energy scheduling constitutes a complex multi-level optimization task regarding the underlying high-dimensional, multi-modal and nonlinear problem structure. The multi-level issue as well as the requirement for model independent algorithm design are substantially supported by appropriate machine learning flexibility models. Generating training sets by digital twins works well for single energy units. Combining training sets from individually modeled energy units, on the other hand, results in folded distributions with unfavorable properties for training. Nevertheless, this happens to be a quite frequent use case, e.g. when an ensemble of distributed energy resources wants to harness the joint flexibility for some control task. A fully decentralized agent-based algorithm is proposed that samples from distributed twins maximizing coverage of flexibility and simultaneously minimizing the discrepancy of the sample by using Ripley’s K measure. Applicability and effectiveness are demonstrated by several simulations using established models for energy unit simulation.

ASJC Scopus Sachgebiete

Zitieren

Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity. / Bremer, Jörg; Gerster, Johannes; Brückner, Birk et al.
Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection: International Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings. Hrsg. / Fernando De La Prieta; Alia El Bolock; Dalila Durães; João Carneiro; Fernando Lopes; Vicente Julian. Cham, 2021. S. 191-202 (Communications in Computer and Information Science; Band 1472 CCIS).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Bremer, J, Gerster, J, Brückner, B, Sarstedt, M & Lehnhoff, S 2021, Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity. in F De La Prieta, A El Bolock, D Durães, J Carneiro, F Lopes & V Julian (Hrsg.), Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection: International Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings. Communications in Computer and Information Science, Bd. 1472 CCIS, Cham, S. 191-202, 19th International Conference on Practical Applications of Agents and Multi-Agent Systems, 6 Okt. 2021. https://doi.org/10.1007/978-3-030-85710-3_16
Bremer, J., Gerster, J., Brückner, B., Sarstedt, M., & Lehnhoff, S. (2021). Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity. In F. De La Prieta, A. El Bolock, D. Durães, J. Carneiro, F. Lopes, & V. Julian (Hrsg.), Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection: International Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings (S. 191-202). (Communications in Computer and Information Science; Band 1472 CCIS).. https://doi.org/10.1007/978-3-030-85710-3_16
Bremer J, Gerster J, Brückner B, Sarstedt M, Lehnhoff S. Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity. in De La Prieta F, El Bolock A, Durães D, Carneiro J, Lopes F, Julian V, Hrsg., Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection: International Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings. Cham. 2021. S. 191-202. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-85710-3_16
Bremer, Jörg ; Gerster, Johannes ; Brückner, Birk et al. / Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity. Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection: International Workshops of PAAMS 2021, Salamanca, Spain, October 6–9, 2021, Proceedings. Hrsg. / Fernando De La Prieta ; Alia El Bolock ; Dalila Durães ; João Carneiro ; Fernando Lopes ; Vicente Julian. Cham, 2021. S. 191-202 (Communications in Computer and Information Science).
Download
@inproceedings{04d04863972542ff869c23087f86a6e0,
title = "Agent-Based Phase Space Sampling of Ensembles using Ripley{\textquoteright}s K for Homogeneity",
abstract = "In future energy systems, decentralized control will require delegation of liabilities to small energy resources. Distributed energy scheduling constitutes a complex multi-level optimization task regarding the underlying high-dimensional, multi-modal and nonlinear problem structure. The multi-level issue as well as the requirement for model independent algorithm design are substantially supported by appropriate machine learning flexibility models. Generating training sets by digital twins works well for single energy units. Combining training sets from individually modeled energy units, on the other hand, results in folded distributions with unfavorable properties for training. Nevertheless, this happens to be a quite frequent use case, e.g. when an ensemble of distributed energy resources wants to harness the joint flexibility for some control task. A fully decentralized agent-based algorithm is proposed that samples from distributed twins maximizing coverage of flexibility and simultaneously minimizing the discrepancy of the sample by using Ripley{\textquoteright}s K measure. Applicability and effectiveness are demonstrated by several simulations using established models for energy unit simulation.",
keywords = "Discrepancy, MAS, Phase space sampling, Scheduling",
author = "J{\"o}rg Bremer and Johannes Gerster and Birk Br{\"u}ckner and Marcel Sarstedt and Sebastian Lehnhoff",
note = "Funding Information: This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) ? 359921210.; 19th International Conference on Practical Applications of Agents and Multi-Agent Systems ; Conference date: 06-10-2021 Through 08-10-2021",
year = "2021",
month = sep,
day = "27",
doi = "10.1007/978-3-030-85710-3_16",
language = "English",
isbn = "978-3-030-85709-7",
series = "Communications in Computer and Information Science",
pages = "191--202",
editor = "{De La Prieta}, Fernando and {El Bolock}, Alia and Dalila Dur{\~a}es and Jo{\~a}o Carneiro and Fernando Lopes and Vicente Julian",
booktitle = "Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection",
url = "https://www.paams.net/",

}

Download

TY - GEN

T1 - Agent-Based Phase Space Sampling of Ensembles using Ripley’s K for Homogeneity

AU - Bremer, Jörg

AU - Gerster, Johannes

AU - Brückner, Birk

AU - Sarstedt, Marcel

AU - Lehnhoff, Sebastian

N1 - Funding Information: This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) ? 359921210.

PY - 2021/9/27

Y1 - 2021/9/27

N2 - In future energy systems, decentralized control will require delegation of liabilities to small energy resources. Distributed energy scheduling constitutes a complex multi-level optimization task regarding the underlying high-dimensional, multi-modal and nonlinear problem structure. The multi-level issue as well as the requirement for model independent algorithm design are substantially supported by appropriate machine learning flexibility models. Generating training sets by digital twins works well for single energy units. Combining training sets from individually modeled energy units, on the other hand, results in folded distributions with unfavorable properties for training. Nevertheless, this happens to be a quite frequent use case, e.g. when an ensemble of distributed energy resources wants to harness the joint flexibility for some control task. A fully decentralized agent-based algorithm is proposed that samples from distributed twins maximizing coverage of flexibility and simultaneously minimizing the discrepancy of the sample by using Ripley’s K measure. Applicability and effectiveness are demonstrated by several simulations using established models for energy unit simulation.

AB - In future energy systems, decentralized control will require delegation of liabilities to small energy resources. Distributed energy scheduling constitutes a complex multi-level optimization task regarding the underlying high-dimensional, multi-modal and nonlinear problem structure. The multi-level issue as well as the requirement for model independent algorithm design are substantially supported by appropriate machine learning flexibility models. Generating training sets by digital twins works well for single energy units. Combining training sets from individually modeled energy units, on the other hand, results in folded distributions with unfavorable properties for training. Nevertheless, this happens to be a quite frequent use case, e.g. when an ensemble of distributed energy resources wants to harness the joint flexibility for some control task. A fully decentralized agent-based algorithm is proposed that samples from distributed twins maximizing coverage of flexibility and simultaneously minimizing the discrepancy of the sample by using Ripley’s K measure. Applicability and effectiveness are demonstrated by several simulations using established models for energy unit simulation.

KW - Discrepancy

KW - MAS

KW - Phase space sampling

KW - Scheduling

UR - http://www.scopus.com/inward/record.url?scp=85116476286&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-85710-3_16

DO - 10.1007/978-3-030-85710-3_16

M3 - Conference contribution

SN - 978-3-030-85709-7

T3 - Communications in Computer and Information Science

SP - 191

EP - 202

BT - Highlights in Practical Applications of Agents, Multi-Agent Systems, and Social Good. The PAAMS Collection

A2 - De La Prieta, Fernando

A2 - El Bolock, Alia

A2 - Durães, Dalila

A2 - Carneiro, João

A2 - Lopes, Fernando

A2 - Julian, Vicente

CY - Cham

T2 - 19th International Conference on Practical Applications of Agents and Multi-Agent Systems

Y2 - 6 October 2021 through 8 October 2021

ER -