A behavior-based intrusion detection technique for smart grid infrastructure

Abstract

A smart grid is a fully automated electricity network, which monitors and controls all its physical environments of electricity infrastructure being able to supply energy in an efficient and reliable way. As the importance of cyber-physical system (CPS) security is growing, various intrusion detection algorithms to protect SCADA system and generation sector have been suggested, whereas there were less consideration on distribution sector. Thus, this paper first highlights the significance of CPS security, especially the availability as the most important factor in smart grid environment. Then this paper classifies various modern intrusion detection system (IDS) techniques for securing smart grid network. In our approach, we propose a novel behavior-based IDS for IEC 61850 protocol using both statistical analysis of traditional network features and specification-based metrics. Finally, we present the attack scenarios and detection methods applicable for IEC 61850-based digital substation in Korean environment.