Enhancing Power Grid Security by Detecting Relay Setting Manipulation Attacks with Machine Learning

As power grids undergo extensive digitalization, they become susceptible to diverse cybersecurity threats, including potentially manipulating protective relay settings. Traditional detection methods struggle to identify these attacks effectively, posing a significant risk of widespread disruptions. This research introduces a machine learning-based approach for detecting and categorizing attacks that manipulate relay settings. The experiments evaluate the efficacy of various algorithms, including Random Forest, XGBoost, and Decision Trees, and Random Forest emerged as the best-performing model, showcasing the capabilities of machine learning to enhance the security of power grid operations.