| Title | Detection of False Data Injection Attacks in smart grids based on cubature Kalman Filtering |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Wang, Zhiwen, Zhang, Qi, Sun, Hongtao, Hu, Jiqiang |
| Conference Name | 2021 33rd Chinese Control and Decision Conference (CCDC) |
| Keywords | attack detection, composability, cosine similarity, Data models, delays, False Data Detection, false data injection attacks, Filtering, Heuristic algorithms, Human Behavior, Power measurement, Prediction algorithms, pubcrawl, resilience, Resiliency, Smart grids, Square root cubature Kalman filter |
| Abstract | The false data injection attacks (FDIAs) in smart grids can offset the power measurement data and it can bypass the traditional bad data detection mechanism. To solve this problem, a new detection mechanism called cosine similarity ratio which is based on the dynamic estimation algorithm of square root cubature Kalman filter (SRCKF) is proposed in this paper. That is, the detection basis is the change of the cosine similarity between the actual measurement and the predictive measurement before and after the attack. When the system is suddenly attacked, the actual measurement will have an abrupt change. However, the predictive measurement will not vary promptly with it owing to the delay of Kalman filter estimation. Consequently, the cosine similarity between the two at this moment has undergone a change. This causes the ratio of the cosine similarity at this moment and that at the initial moment to fluctuate considerably compared to safe operation. If the detection threshold is triggered, the system will be judged to be under attack. Finally, the standard IEEE-14bus test system is used for simulation experiments to verify the effectiveness of the proposed detection method. |
| DOI | 10.1109/CCDC52312.2021.9601667 |
| Citation Key | wang_detection_2021 |
Detection of False Data Injection Attacks in smart grids based on cubature Kalman Filtering