Detection of False Data Injection Attacks in Distribution System State Estimation

To establish a well-grounded operation of a utility grid, accurate measurements of system state is needed. The State Estimation is essential for monitoring and controlling the operation of power system. Over the years weighted least square state estimation has been dominant. This methodology was originally developed for transmission systems and proves to be inefficient due to the complexity of the distribution systems. Development of Smart Grids has created a substantial impact in the monitoring of distribution systems. Therefore, a branch current based distribution system state estimation is developed. Recent cyber-attacks on power systems exhibits the loopholes present in this method. Carefully knitted stealth bad data injection attack can penetrate the system and surpass the state estimation. Thereby remaining undetected. By this way an attacker can launch a successful cyber-attack. A detection strategy is developed that traces the dynamics in measurement variations and detects the false data injection attacks. The proposed algorithm is tested and validated on a simulated case study using IEEE 33 bus distribution system.