An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold

Yufeng Wang, Wanjiao Shi, Qun Jin, Jianhua Ma

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Smart grids are vulnerable to cyber-attacks, which can cause significant damage and huge economic losses. Generally, state estimation (SE) is used to observe the operation of the grid. State estimation of the grid is vulnerable to false data injection attack (FDIA), so diagnosing this type of malicious attack has a major impact on ensuring reliable operation of the power system. In this paper, we present an effective FDIA detection method based on residual recurrent neural network (R2N2) prediction model and adaptive judgment threshold. Specifically, considering the data contains both linear and nonlinear components, the R2N2 model divides the prediction process into two parts: The first part uses the linear model to fit the state data; the second part predicts the nonlinearity of the residuals of the linear prediction model. The adaptive judgment threshold is inferred through fitting the Weibull distribution with the sum of squared errors between the predicted values and observed values. The thorough simulation results demonstrate that our scheme performs better than other prediction based FDIA detection schemes.

Original languageEnglish
Title of host publicationProceedings - IEEE International Conference on Energy Internet, ICEI 2019
EditorsGuokai Wu, Jiye Wang, Qinliang Tan
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages499-504
Number of pages6
ISBN (Electronic)9781728114934
DOIs
Publication statusPublished - 2019 May
Event3rd IEEE International Conference on Energy Internet, ICEI 2019 - Nanjing, China
Duration: 2019 May 272019 May 31

Publication series

NameProceedings - IEEE International Conference on Energy Internet, ICEI 2019

Conference

Conference3rd IEEE International Conference on Energy Internet, ICEI 2019
CountryChina
CityNanjing
Period19/5/2719/5/31

Fingerprint

Adaptive Threshold
Smart Grid
Recurrent neural networks
Recurrent Neural Networks
Attack
State estimation
Injection
State Estimation
Prediction Model
Linear Model
Weibull distribution
Grid
Linear Prediction
Prediction
Weibull Distribution
Power System
Network Model
Divides
Damage
Economics

Keywords

  • Adaptive detection threshold
  • False data injection attack
  • Residual recurrent neural network
  • State estimation

ASJC Scopus subject areas

  • Control and Optimization
  • Computer Networks and Communications
  • Energy Engineering and Power Technology
  • Renewable Energy, Sustainability and the Environment
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality

Cite this

Wang, Y., Shi, W., Jin, Q., & Ma, J. (2019). An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold. In G. Wu, J. Wang, & Q. Tan (Eds.), Proceedings - IEEE International Conference on Energy Internet, ICEI 2019 (pp. 499-504). [8791377] (Proceedings - IEEE International Conference on Energy Internet, ICEI 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICEI.2019.00094

An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold. / Wang, Yufeng; Shi, Wanjiao; Jin, Qun; Ma, Jianhua.

Proceedings - IEEE International Conference on Energy Internet, ICEI 2019. ed. / Guokai Wu; Jiye Wang; Qinliang Tan. Institute of Electrical and Electronics Engineers Inc., 2019. p. 499-504 8791377 (Proceedings - IEEE International Conference on Energy Internet, ICEI 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, Y, Shi, W, Jin, Q & Ma, J 2019, An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold. in G Wu, J Wang & Q Tan (eds), Proceedings - IEEE International Conference on Energy Internet, ICEI 2019., 8791377, Proceedings - IEEE International Conference on Energy Internet, ICEI 2019, Institute of Electrical and Electronics Engineers Inc., pp. 499-504, 3rd IEEE International Conference on Energy Internet, ICEI 2019, Nanjing, China, 19/5/27. https://doi.org/10.1109/ICEI.2019.00094
Wang Y, Shi W, Jin Q, Ma J. An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold. In Wu G, Wang J, Tan Q, editors, Proceedings - IEEE International Conference on Energy Internet, ICEI 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 499-504. 8791377. (Proceedings - IEEE International Conference on Energy Internet, ICEI 2019). https://doi.org/10.1109/ICEI.2019.00094
Wang, Yufeng ; Shi, Wanjiao ; Jin, Qun ; Ma, Jianhua. / An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold. Proceedings - IEEE International Conference on Energy Internet, ICEI 2019. editor / Guokai Wu ; Jiye Wang ; Qinliang Tan. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 499-504 (Proceedings - IEEE International Conference on Energy Internet, ICEI 2019).
@inproceedings{55d7612c377e4728b8f2db0309957fd5,
title = "An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold",
abstract = "Smart grids are vulnerable to cyber-attacks, which can cause significant damage and huge economic losses. Generally, state estimation (SE) is used to observe the operation of the grid. State estimation of the grid is vulnerable to false data injection attack (FDIA), so diagnosing this type of malicious attack has a major impact on ensuring reliable operation of the power system. In this paper, we present an effective FDIA detection method based on residual recurrent neural network (R2N2) prediction model and adaptive judgment threshold. Specifically, considering the data contains both linear and nonlinear components, the R2N2 model divides the prediction process into two parts: The first part uses the linear model to fit the state data; the second part predicts the nonlinearity of the residuals of the linear prediction model. The adaptive judgment threshold is inferred through fitting the Weibull distribution with the sum of squared errors between the predicted values and observed values. The thorough simulation results demonstrate that our scheme performs better than other prediction based FDIA detection schemes.",
keywords = "Adaptive detection threshold, False data injection attack, Residual recurrent neural network, State estimation",
author = "Yufeng Wang and Wanjiao Shi and Qun Jin and Jianhua Ma",
year = "2019",
month = "5",
doi = "10.1109/ICEI.2019.00094",
language = "English",
series = "Proceedings - IEEE International Conference on Energy Internet, ICEI 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "499--504",
editor = "Guokai Wu and Jiye Wang and Qinliang Tan",
booktitle = "Proceedings - IEEE International Conference on Energy Internet, ICEI 2019",

}

TY - GEN

T1 - An accurate false data detection in smart grid based on residual recurrent neural network and adaptive threshold

AU - Wang, Yufeng

AU - Shi, Wanjiao

AU - Jin, Qun

AU - Ma, Jianhua

PY - 2019/5

Y1 - 2019/5

N2 - Smart grids are vulnerable to cyber-attacks, which can cause significant damage and huge economic losses. Generally, state estimation (SE) is used to observe the operation of the grid. State estimation of the grid is vulnerable to false data injection attack (FDIA), so diagnosing this type of malicious attack has a major impact on ensuring reliable operation of the power system. In this paper, we present an effective FDIA detection method based on residual recurrent neural network (R2N2) prediction model and adaptive judgment threshold. Specifically, considering the data contains both linear and nonlinear components, the R2N2 model divides the prediction process into two parts: The first part uses the linear model to fit the state data; the second part predicts the nonlinearity of the residuals of the linear prediction model. The adaptive judgment threshold is inferred through fitting the Weibull distribution with the sum of squared errors between the predicted values and observed values. The thorough simulation results demonstrate that our scheme performs better than other prediction based FDIA detection schemes.

AB - Smart grids are vulnerable to cyber-attacks, which can cause significant damage and huge economic losses. Generally, state estimation (SE) is used to observe the operation of the grid. State estimation of the grid is vulnerable to false data injection attack (FDIA), so diagnosing this type of malicious attack has a major impact on ensuring reliable operation of the power system. In this paper, we present an effective FDIA detection method based on residual recurrent neural network (R2N2) prediction model and adaptive judgment threshold. Specifically, considering the data contains both linear and nonlinear components, the R2N2 model divides the prediction process into two parts: The first part uses the linear model to fit the state data; the second part predicts the nonlinearity of the residuals of the linear prediction model. The adaptive judgment threshold is inferred through fitting the Weibull distribution with the sum of squared errors between the predicted values and observed values. The thorough simulation results demonstrate that our scheme performs better than other prediction based FDIA detection schemes.

KW - Adaptive detection threshold

KW - False data injection attack

KW - Residual recurrent neural network

KW - State estimation

UR - http://www.scopus.com/inward/record.url?scp=85071529371&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071529371&partnerID=8YFLogxK

U2 - 10.1109/ICEI.2019.00094

DO - 10.1109/ICEI.2019.00094

M3 - Conference contribution

AN - SCOPUS:85071529371

T3 - Proceedings - IEEE International Conference on Energy Internet, ICEI 2019

SP - 499

EP - 504

BT - Proceedings - IEEE International Conference on Energy Internet, ICEI 2019

A2 - Wu, Guokai

A2 - Wang, Jiye

A2 - Tan, Qinliang

PB - Institute of Electrical and Electronics Engineers Inc.

ER -