Cyberattack Classification in Smart Grid Distribution Substations using a Novel Ensemble Bagging Learning Technique

Smart grid incorporates the communication networking that enables the exchange of information among the monitoring and controlling devices. Such incorporation of the communication networking into the electricity grid infrastructure poses the risk of cyberattacks that target the critical assets within such an infrastructure. Most of the existing research focuses on the detection of such cyberattacks but without identifying the type of the attacks. This can result in overlooked threats and misdirected the necessary countermeasures. Recognizing the attack's type is essential for timely responses and strategic planning against future threats, thereby enhancing the resilience of the smart grid. In this paper, a Fine Tree Bagging-based Ensemble Learning (FTBE) technique is proposed to detect and classify the different types of cyberattacks and power quality disturbances. The salient features of the attacks' types are highlighted, which helps in identifying the types of the attack following the detection process.