Deep Neural Network-Based Stealthy False Data Injection Attack Detection on Distributed Energy Resources Integrated Systems

State estimation is a critical process for ensuring the secure and reliable operation of power systems by determining the operating state of a system based on available measurements. Recent studies indicate that the state estimation process can be compromised by False Data Injection Attacks (FDIA). These attacks involve injecting attack vectors into compromised measurements to evade bad data detection methods. While conventional state estimation is already susceptible to such attacks, the increasing penetration of distributed energy resources (DERs) has exacerbated the system's vulnerability to cyber-attacks. In this paper, we propose a deep learning-based FDIA detection method to identify cyber-attacks in power systems with a high penetration rate of DERs. We compare this method with widely-used classification algorithms for detecting anomalies in state estimation and measurements. The proposed approaches are evaluated using historical load data from the New York Independent System Operator (NYISO) on two IEEE test systems: the 30-bus and 57-bus systems. The methods are tested under various DER contributions and noise levels. Results demonstrate the efficacy of the proposed methods in detecting FDIAs in power systems with varying levels of DER penetration.