Energy Big Data Analytics from a Cybersecurity Perspective

Decorative dot pattern

    The project aims to detect cyber attacks, especially false data injection (FDI) attacks, on smart grid infrastructure via data analytics. It includes monitoring the smart grid operation via deep learning model and detection of the FDI attack from the big smart meter readings received for the smart grid state estimation that is used for power control.

    Problem statement

    A recent cyber attack (false data injection attack FDI) has been found that can penetrate the widely used bad data detection mechanism in smart grid. FDI attack modifies the smart meter readings such that the modified data satisfy the physical power flow law which cannot be detected by mechanism the bad data detection. We use deep learning model to detect the anomaly of the smart grid state.

    Application and Impact

    Smart grid uses smart readings to perform power scheduling and real-time control. FDI attacks can disrupt the operation of the smart grid, leading to the crash of the infrastructure.