Intrusion Detection in Cyber-physical Layer of Smart Grid using Intelligent Loop Based Artificial Neural Network Technique

This paper, proposes an Intelligent Loop Based Artificial Neural Network (ILANN) based detection technique for the detection of cyber intrusion in a smart grid against False Data Injection Attack (FDIA). This method compares the deviation of a system with the equipment load profile present on the system node(s) and any deviation from predefined values generates an alarm. Every 2 milliseconds (ms) the data obtained by the measurement is passed through the attack detection system, in case if the deviation is continuously for 5 measurement cycles i.e. for 10 ms and it does not match with the load combination the operator will get the first alert alarm. In case the deviation is not fixed after 8 measurement cycles then the system alerts the control centre. FDI attack is used by attackers to affect the healthy operation of the smart grid. Using FDI the hackers can permanently damage many power system equipment's which may lead to higher fixing costs. The result and analysis of the proposed cyber detection approach help operator and control centre to identify cyber intrusion in the smart grid scenario. The method is used to detect a cyberattack on IEEE-9 Bus test system using MATLAB software.