Detection of False Data Injection Attacks Using the Autoencoder Approach

Chenguang Wang; Simon Tindemans; Kaikai Pan; Peter Palensky

doi:10.1109/PMAPS47429.2020.9183526

Detection of False Data Injection Attacks Using the Autoencoder Approach

Chenguang Wang, Simon Tindemans, Kaikai Pan, Peter Palensky

Intelligent Electrical Power Grids

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

9 Citations (Scopus)

14 Downloads (Pure)

Abstract

State estimation is of considerable significance for the power system operation and control. However, well-designed false data injection attacks can utilize blind spots in conventional residual-based bad data detection methods to manipulate measurements in a coordinated manner and thus affect the secure operation and economic dispatch of grids. In this paper, we propose a detection approach based on an autoencoder neural network. By training the network on the dependencies intrinsic in ‘normal’ operation data, it effectively overcomes the challenge of unbalanced training data that is inherent in power system attack detection. To evaluate the detection performance of the proposed mechanism, we conduct a series of experiments on the IEEE 118-bus power system. The experiments demonstrate that the proposed autoencoder detector displays robust detection performance under a variety of attack scenarios.

Original language	English
Title of host publication	2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS)
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-7281-2822-1
DOIs	https://doi.org/10.1109/PMAPS47429.2020.9183526
Publication status	Published - 2020
Event	2020 International Conference on Probabilistic Methods Applied to Power Systems - , Belgium Duration: 18 Aug 2020 → 21 Aug 2020

Conference

Conference	2020 International Conference on Probabilistic Methods Applied to Power Systems
Abbreviated title	PMAPS 2020
Country/Territory	Belgium
Period	18/08/20 → 21/08/20
Other	Virtual/online event due to COVID-19

Bibliographical note

Virtual/online event due to COVID-19
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

nomaly detection
autoencoder
false data in-jection attack
unbalanced training data
machine learning

Access to Document

10.1109/PMAPS47429.2020.9183526

Detection_of_False_Data_Injection_Attacks_Using_the_Autoencoder_ApproachFinal published version, 298 KB

Cite this

@inproceedings{ca209c17572946858d09522dbd9f1a77,

title = "Detection of False Data Injection Attacks Using the Autoencoder Approach",

abstract = "State estimation is of considerable significance for the power system operation and control. However, well-designed false data injection attacks can utilize blind spots in conventional residual-based bad data detection methods to manipulate measurements in a coordinated manner and thus affect the secure operation and economic dispatch of grids. In this paper, we propose a detection approach based on an autoencoder neural network. By training the network on the dependencies intrinsic in {\textquoteleft}normal{\textquoteright} operation data, it effectively overcomes the challenge of unbalanced training data that is inherent in power system attack detection. To evaluate the detection performance of the proposed mechanism, we conduct a series of experiments on the IEEE 118-bus power system. The experiments demonstrate that the proposed autoencoder detector displays robust detection performance under a variety of attack scenarios.",

keywords = "nomaly detection, autoencoder, false data in-jection attack, unbalanced training data, machine learning",

author = "Chenguang Wang and Simon Tindemans and Kaikai Pan and Peter Palensky",

note = "Virtual/online event due to COVID-19 Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – 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. ; 2020 International Conference on Probabilistic Methods Applied to Power Systems, PMAPS 2020 ; Conference date: 18-08-2020 Through 21-08-2020",

year = "2020",

doi = "10.1109/PMAPS47429.2020.9183526",

language = "English",

booktitle = "2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS)",

publisher = "IEEE ",

address = "United States",

}

Wang, C , Tindemans, S , Pan, K & Palensky, P 2020, Detection of False Data Injection Attacks Using the Autoencoder Approach. in 2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS). IEEE , 2020 International Conference on Probabilistic Methods Applied to Power Systems, Belgium, 18/08/20. https://doi.org/10.1109/PMAPS47429.2020.9183526

Detection of False Data Injection Attacks Using the Autoencoder Approach. / Wang, Chenguang ; Tindemans, Simon ; Pan, Kaikai ; Palensky, Peter.

2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS). IEEE , 2020.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Detection of False Data Injection Attacks Using the Autoencoder Approach

AU - Wang, Chenguang

AU - Tindemans, Simon

AU - Pan, Kaikai

AU - Palensky, Peter

N1 - Virtual/online event due to COVID-19 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 - 2020

Y1 - 2020

N2 - State estimation is of considerable significance for the power system operation and control. However, well-designed false data injection attacks can utilize blind spots in conventional residual-based bad data detection methods to manipulate measurements in a coordinated manner and thus affect the secure operation and economic dispatch of grids. In this paper, we propose a detection approach based on an autoencoder neural network. By training the network on the dependencies intrinsic in ‘normal’ operation data, it effectively overcomes the challenge of unbalanced training data that is inherent in power system attack detection. To evaluate the detection performance of the proposed mechanism, we conduct a series of experiments on the IEEE 118-bus power system. The experiments demonstrate that the proposed autoencoder detector displays robust detection performance under a variety of attack scenarios.

AB - State estimation is of considerable significance for the power system operation and control. However, well-designed false data injection attacks can utilize blind spots in conventional residual-based bad data detection methods to manipulate measurements in a coordinated manner and thus affect the secure operation and economic dispatch of grids. In this paper, we propose a detection approach based on an autoencoder neural network. By training the network on the dependencies intrinsic in ‘normal’ operation data, it effectively overcomes the challenge of unbalanced training data that is inherent in power system attack detection. To evaluate the detection performance of the proposed mechanism, we conduct a series of experiments on the IEEE 118-bus power system. The experiments demonstrate that the proposed autoencoder detector displays robust detection performance under a variety of attack scenarios.

KW - nomaly detection

KW - autoencoder

KW - false data in-jection attack

KW - unbalanced training data

KW - machine learning

U2 - 10.1109/PMAPS47429.2020.9183526

DO - 10.1109/PMAPS47429.2020.9183526

M3 - Conference contribution

BT - 2020 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS)

PB - IEEE

T2 - 2020 International Conference on Probabilistic Methods Applied to Power Systems

Y2 - 18 August 2020 through 21 August 2020

ER -

Detection of False Data Injection Attacks Using the Autoencoder Approach

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this