Research on False Data Injection Attack Detection of S mart Grid Based on Machine Learning

Abstract

Abstract The safe and reliable operation of power systems is an important guarantee for the healthy development of the national economy. Industry and people’s lives are inseparable from electricity, so the safety and reliability of electricity supply is very important. The sudden interruption of power supply will not only bring serious economic losses, but also seriously affect people’s normal lives and even endanger social stability. False data injection attack (FDIAs) are a new type of power system network attack method. FDIAs are a new type of power system network attack method. It can successfully bypass the bad data detection mechanism, offset the power measurement data, and mislead the control center under extremely subtle conditions. Therefore, it poses a very serious threat to the stable operation of the power system. Therefore, this article first analyzes the principle of false data injection attacks, in order to provide a theoretical basis for subsequent attack detection. Then this paper constructs a detection method based on extreme learning from the perspective of optimizing learning efficiency. Based on the IEEE14-bus standard test system, the method is verified through simulation, which shows the feasibility of this method, which provides a direction for building a safe and stable smart grid.