Abstract
Smart grids have become susceptible to cyber-attacks, being one of the most diversified cyber–physical systems. Measurements collected by the supervisory control and data acquisition system can be compromised by a smart hacker, who can cheat a bad-data detector during state estimation by injecting biased values into the sensor-collected measurements. This may result in false control decisions, compromising the security of the smart grid, and leading to financial losses, power network disruptions, or a combination of both. To overcome these problems, we propose a novel approach to cyber-attacks detection, based on an extremely randomized trees algorithm and kernel principal component analysis for dimensionality reduction. A performance evaluation of the proposed scheme is done by using the standard IEEE 57-bus and 118-bus systems. Numerical results show that the proposed scheme outperforms state-of-art approaches while improving the accuracy in detection of stealth cyber-attacks in smart-grid measurements.
Highlights
The notion of smart grids (SGs) is realized by modern computing and bi-directional communications systems being combined with the typical electrical power grid
In order to solve the computational complexity created by a high-dimensional space in large-sized power systems, we apply the Kernel PCA (KPCA) technique to transform the data into a lower-dimensional space
The dataset is composed of historical active power flow measurements and active power injections into the buses, which were collected at the power control center (PCC) of the power network
Summary
The notion of smart grids (SGs) is realized by modern computing and bi-directional communications systems being combined with the typical electrical power grid. Ozay et al [10] provided a comparison study of several ML algorithms, such as the SVM, AdaBoost, and perceptron, for attack detection in SGs, in which the ML-based approaches achieved a higher performance, compared with state vector estimation methods. The computational cost at the PCC is reduced, since we decrease the number of features, where a fast and efficient classification algorithm is needed to keep the computational cost of the entire process as low as possible Based on this objective, we propose an extremely randomized trees (ExtraTrees)-based approach to detect SCA attacks in SGs. The Extra-Trees algorithm is an ensemble method characterized by being computationally efficient and providing high accuracy [15], where the strength of the randomization helps to achieve a greater reduction in the variance, compared with other ensemble methods like random forest or AdaBoost.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
