Automated tiered Detection of Cyber-Attacks in Operational Distribution Networks

The increasing dependence on digital technologies within power grids brings forth a pressing concern: the vulnerability to cyber-attacks poses a significant threat to its security and reliability. In response, this research proposes an innovative approach leveraging ensemble learning techniques to predict and mitigate cyber-attacks in Smart Grids. By combining the strengths of diverse models such as Extra Trees, XGBoost, Random Forest, and Bagging, the ensemble model aims to bolster attack detection accuracy and robustness. Through real-time monitoring, anomaly detection, and rapid response mechanisms, the framework enhances the resilience of Smart Grids against evolving cyber threats. The research focuses on refining a predictive model capable of distinguishing legitimate data from malicious injections, thus ensuring the ongoing security and reliability of critical energy infrastructure. This innovative ensemble learning approach represents a promising strategy to fortify modern power grids, safeguarding against sophisticated cyber threats and ensuring continued operational integrity.