On Analysis of Effectiveness of Ensemble, Distance and Tree based Methods for Secure Power Systems

The inclusion of information and communication technology in power systems has introduced us with smart power systems. As countries around the globe are moving towards building a highly power-driven society, this has largely brought the power system at the epicenter of cyber-attacks. In this scenario the rule-based systems are not efficient enough to provide smart solutions to mitigate the problem of cyber-attacks. Therefore, this paper works as a starting point to investigate the potential to build an Artificially Intelligent Power System Intrusion Detection mechanism. The authors experiment with machine learning models to build a system capable of mitigating the Intrusion attacks on power systems. The empirical results indicate that Ensemble-method is better than Distance-based or Tree-based method for the problem.