| Title | False Data Injection Attacks in Smart Grid Using Gaussian Mixture Model |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Wang, Zhiwen, Hu, Jiqiang, Sun, Hongtao |
| Conference Name | 2020 16th International Conference on Control, Automation, Robotics and Vision (ICARCV) |
| Date Published | dec |
| Keywords | compositionality, Density measurement, Human Behavior, human factors, Mathematical model, Power measurement, pubcrawl, resilience, Resiliency, Smart Grid Sensors, Smart grids, state estimation, Time measurement, Weight measurement |
| Abstract | The application of network technology and high-tech equipment in power systems has increased the degree of grid intelligence, and malicious attacks on smart grids have also increased year by year. The wrong data injection attack launched by the attacker will destroy the integrity of the data by changing the data of the sensor and controller, which will lead to the wrong decision of the control system and even paralyze the power transmission network. This paper uses the measured values of smart grid sensors as samples, analyzes the attack vectors maliciously injected by attackers and the statistical characteristics of system data, and proposes a false data injection attack detection strategy. It is considered that the measured values of sensors have spatial distribution characteristics, the Gaussian mixture model of grid node feature vectors is obtained by training sample values, the test measurement values are input into the Gaussian mixture model, and the knowledge of clustering is used to detect whether the power grid is malicious data attacks. The power supplies of IEEE-18 and IEEE-30 simulation systems was tested, and the influence of the system statistical measurement characteristics on the detection accuracy was analyzed. The results show that the proposed strategy has better detection performance than the support vector machine method. |
| DOI | 10.1109/ICARCV50220.2020.9305398 |
| Citation Key | wang_false_2020 |
False Data Injection Attacks in Smart Grid Using Gaussian Mixture Model