Study of Cyber Threats and its Challenges in LFC and its Mitigation via GDA based PID Controller

The interconnected power system has evolved rapidly in the modern era. This results in a new global infrastructure for distributed systems from locally distributed systems. This established a network-based cyber-physical system (NCPS). Many crucial operations need to be performed for a better quality of power. Among them, frequency stabilisation is a significant concern against uncertainties, random load changes, etc. The fluctuations in frequency can cause a severe issue for the whole power system, so load frequency control (LFC) is an important task to be performed. Due to NCPS, LFC functioning is highly vulnerable to cyber threats. Different kind of attacks can affect LFC operations, such as denial of service (DoS), false data injection (FDI), time delay attacks (Tda), etc. The above problems motivate us to study and analyse the cyber threat issues in NCPS during LFC operations. In view of the above, various existing cyber-attacks were discussed. Further, various detection approaches using machine learning (ML) were studied. In order to show the application of the ML-based control technique, a new PID tuning approach is proposed, which is based on the gradient descent algorithm (GDA) for the mitigation of cyber-attacks, especially for the single area power system model.