Identification scheme of false data injection attack based on deep learning algorithms for smart grids

Abstract

This paper presents the artificial intelligence (AI) techniques based on the deep learning algorithms to diagnose false data injection (FDI) attacks to smart grids with the measurement data in real-time. The power and data flow between end-user consumers and all components of the advanced metering infrastructure (AMI) and supervisory control and data acquisition (SCADA) system in the SG is bidirectional flow by advanced communication networks. For all the advantages of the SG come with, they remain at risk to a series of many potential threats and ongoing attacks. The conditional-deep-belief-network (CDBN) architecture is employed to un-observable FDI attacks which pass the state-vector-estimator (SVE) mechanisms. The IEEE 118 bus, and IEEE 300 bus power system have been used to evaluate our detection scheme. Finally, the suggested CDBN scheme is compared with other detection such as artificial neural network (ANN) and support vector machine (SVM). It is observed that the simulation result shows that suggested detection methods can attain a high accuracy of unobservable FDI attacks.