False Data Injection Attack Detection in a Power Grid Using RNN

Cyber attacks on Cyber Physical Systems (CPSs), especially on those critical infrastructures poses severe threat on the public security. Among them, a special kind of attack, False Data Injection (FDI), can bypass the surveillance of state-estimation-based bad data detection mechanism silently. In this paper, we exploited the strong ability of Recurrent Neural Network (RNN) on time-series prediction to recognize the potential compromised measurements. It makes our proposed method practicable in real-world scenario that no labeled data is required during all stages of algorithm. An experiment on IEEE-14 bus test system is conducted and shows a promising result that our proposed method is able to detect FDI attack with high precision and high recall.

[1]  A. G. Expósito,et al.  Power system state estimation : theory and implementation , 2004 .

[2]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[3]  Ran Li,et al.  Deep Learning for Household Load Forecasting—A Novel Pooling Deep RNN , 2018, IEEE Transactions on Smart Grid.

[4]  Jin Wei,et al.  Real-Time Detection of False Data Injection Attacks in Smart Grid: A Deep Learning-Based Intelligent Mechanism , 2017, IEEE Transactions on Smart Grid.

[5]  Zhu Han,et al.  Detecting False Data Injection Attacks on Power Grid by Sparse Optimization , 2014, IEEE Transactions on Smart Grid.

[6]  Lovekesh Vig,et al.  Long Short Term Memory Networks for Anomaly Detection in Time Series , 2015, ESANN.

[7]  Ali Reza Seifi,et al.  A statistical unsupervised method against false data injection attacks: A visualization-based approach , 2017, Expert Syst. Appl..

[8]  Peng Ning,et al.  False data injection attacks against state estimation in electric power grids , 2011, TSEC.

[9]  Zhao Yang Dong,et al.  The 2015 Ukraine Blackout: Implications for False Data Injection Attacks , 2017, IEEE Transactions on Power Systems.

[10]  Adnan Anwar,et al.  Vulnerabilities of Smart Grid State Estimation against False Data Injection Attack , 2014, ArXiv.

[11]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[12]  Donghua Zhou,et al.  Two-Channel False Data Injection Attacks Against Output Tracking Control of Networked Systems , 2016, IEEE Transactions on Industrial Electronics.

[13]  H. Vincent Poor,et al.  Machine Learning Methods for Attack Detection in the Smart Grid , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[14]  Pavel Filonov,et al.  RNN-based Early Cyber-Attack Detection for the Tennessee Eastman Process , 2017, ArXiv.

[15]  Pavel Filonov,et al.  Multivariate Industrial Time Series with Cyber-Attack Simulation: Fault Detection Using an LSTM-based Predictive Data Model , 2016, ArXiv.

[16]  Fei Hu,et al.  Detection of Faults and Attacks Including False Data Injection Attack in Smart Grid Using Kalman Filter , 2014, IEEE Transactions on Control of Network Systems.

[17]  R D Zimmerman,et al.  MATPOWER: Steady-State Operations, Planning, and Analysis Tools for Power Systems Research and Education , 2011, IEEE Transactions on Power Systems.