A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology

Yigu Liu; Haiwei Xie; Alfan  Presekal; Alexandru Stefanov; Peter Palensky

doi:10.1109/TSG.2023.3304134

A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology

Yigu Liu^*, Haiwei Xie, Alfan Presekal, Alexandru Stefanov, Peter Palensky

^*Corresponding author for this work

Research output: Contribution to journal › Article › Scientific › peer-review

2 Citations (Scopus)

24 Downloads (Pure)

Abstract

Synthetic networks aim at generating realistic projections of real-world networks while concealing the actual system information. This paper proposes a scalable and effective approach based on graph neural networks (GNN) to generate synthetic topologies of Cyber-Physical power Systems (CPS) with realistic network feature distribution. In order to comprehensively capture the characteristics of real CPS networks, we propose a generative model, namely Graph-CPS, based on graph variational autoencoder and graph recurrent neural networks. The method hides the sensitive topological information while maintaining the similar feature distribution of the real networks. We used multiple power and communication networks to prove and assess the effectiveness of the proposed method with experimental results.

Original language	English
Pages (from-to)	4968-4971
Number of pages	4
Journal	IEEE Transactions on Smart Grid
Volume	14
Issue number	6
DOIs	https://doi.org/10.1109/TSG.2023.3304134
Publication status	Published - 2023

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Cyber-physical systems
graph neural networks
synthetic networks

Access to Document

10.1109/TSG.2023.3304134

A_GNN-Based_Generative_Model_for_Generating_Synthetic_Cyber-Physical_Power_System_TopologyFinal published version, 1.27 MB

Cite this

@article{971c7c1abc35499ebd2a08d362760d55,

title = "A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology",

abstract = "Synthetic networks aim at generating realistic projections of real-world networks while concealing the actual system information. This paper proposes a scalable and effective approach based on graph neural networks (GNN) to generate synthetic topologies of Cyber-Physical power Systems (CPS) with realistic network feature distribution. In order to comprehensively capture the characteristics of real CPS networks, we propose a generative model, namely Graph-CPS, based on graph variational autoencoder and graph recurrent neural networks. The method hides the sensitive topological information while maintaining the similar feature distribution of the real networks. We used multiple power and communication networks to prove and assess the effectiveness of the proposed method with experimental results.",

keywords = "Cyber-physical systems, graph neural networks, synthetic networks",

author = "Yigu Liu and Haiwei Xie and Alfan Presekal and Alexandru Stefanov and Peter Palensky",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.",

year = "2023",

doi = "10.1109/TSG.2023.3304134",

language = "English",

volume = "14",

pages = "4968--4971",

journal = "IEEE Transactions on Smart Grid",

issn = "1949-3061",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

number = "6",

}

TY - JOUR

T1 - A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology

AU - Liu, Yigu

AU - Xie, Haiwei

AU - Presekal, Alfan

AU - Stefanov, Alexandru

AU - Palensky, Peter

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2023

Y1 - 2023

N2 - Synthetic networks aim at generating realistic projections of real-world networks while concealing the actual system information. This paper proposes a scalable and effective approach based on graph neural networks (GNN) to generate synthetic topologies of Cyber-Physical power Systems (CPS) with realistic network feature distribution. In order to comprehensively capture the characteristics of real CPS networks, we propose a generative model, namely Graph-CPS, based on graph variational autoencoder and graph recurrent neural networks. The method hides the sensitive topological information while maintaining the similar feature distribution of the real networks. We used multiple power and communication networks to prove and assess the effectiveness of the proposed method with experimental results.

AB - Synthetic networks aim at generating realistic projections of real-world networks while concealing the actual system information. This paper proposes a scalable and effective approach based on graph neural networks (GNN) to generate synthetic topologies of Cyber-Physical power Systems (CPS) with realistic network feature distribution. In order to comprehensively capture the characteristics of real CPS networks, we propose a generative model, namely Graph-CPS, based on graph variational autoencoder and graph recurrent neural networks. The method hides the sensitive topological information while maintaining the similar feature distribution of the real networks. We used multiple power and communication networks to prove and assess the effectiveness of the proposed method with experimental results.

KW - Cyber-physical systems

KW - graph neural networks

KW - synthetic networks

UR - http://www.scopus.com/inward/record.url?scp=85167795984&partnerID=8YFLogxK

U2 - 10.1109/TSG.2023.3304134

DO - 10.1109/TSG.2023.3304134

M3 - Article

SN - 1949-3061

VL - 14

SP - 4968

EP - 4971

JO - IEEE Transactions on Smart Grid

JF - IEEE Transactions on Smart Grid

IS - 6

ER -

A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this