Detection of FDI Attacks on Voltage Regulation of PV-Integrated Distribution Grids Using Machine Learning Methods

The integration of photovoltaic (PV) panels in power distribution systems (PDSs) has provided the grid with the capability to regulate voltage through the injection or absorption of reactive power. The deployment of information and communication technologies (ICTs), which is required for this voltage regulation scheme, has made the PDS prone to a variety of cyber attacks, e.g., false data injection (FDI) attacks. To counter these attacks, this paper proposes a data-driven detection framework to identify FDI attacks against voltage regulation of PV-integrated PDS. To regulate the voltage at a desired location, e.g., the point of common coupling (PCC), the voltage measurements are sent to a centralized controller and the calculated control signals are transmitted back to PVs to be added to their internal control schemes. During this transmission of data, an attacker manipulates the measurement data and launches an FDI attack leading to an unacceptable voltage profile and operation of protection systems. To detect these attacks, a machine learning (ML)-based framework based on support vector machine (SVM) is developed in this research. In this regard, a dataset of different operating points, e.g., loading conditions, is collected to train this supervised framework. The performance of the trained framework for attack detection has been compared with other supervised and unsupervised ML-based techniques in the case of FDI attacks against modified IEEE 33-bus PDS. The obtained results demonstrate the superior performance of the proposed framework in detecting FDI attacks.