Improved-ELM method for detecting false data attack in smart grid. (October 2017)

• Record Type: Journal Article
• Authors: Yang, Liqun; Li, Yuancheng; Li, Zhoujun
• Is Part Of: International journal of electrical power & energy systems. Volume 91(2017)
• Page Start: 183
• Page End: 191
• Publication Date: 2017-10
• Subjects: Smart grid -- False data injection attack -- Dimension reduction -- Extreme learning machine (ELM); Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2017.03.011
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: In this paper, the extreme learning machine is optimized by ABC-DE algorithm for detecting unobservable false data injection attack. We use a deep learning method-Autoencoder to reduce the dimension of the measurements. We conduct lots of experiments to verify the applicability of the proposed methods. This paper firstly constructs attack vector, then simulate the normal and false measurements. Using the generated active power measurement data as experimental data. The deep learning algorithm is used to reduce the dimension of data, which not only normalizes the experiment data, but also makes data show the good divisibility in low dimension space. In order to improve the performance of detection methods, we blend the thought of differential evolution into artificial bee colony algorithm. Using ABC-DE to optimize the weights and other parameters of ELM classifier, and verifying the proposed detection method is superior to other machine learning methods. This paper opens a new era of using ELM with deep learning method to detect the false data injection attack, and the relevant work will be furtherly carried out in the future. Abstract: Power grid is a complex system which closely links the power generation and power consumer through transmission and distribution networks. With the development of smart grid, smart grid is more open to external communication systems, it also has exposed some problems in the network attacks. A new false data injection attack (called the unobservable attack) that can bypass the traditional BDD and inject random errors into state estimation. We propose an improved extreme learning machine (ELM) for attack detection. The artificial bee colony (ABC) incorporates the thought of differential evolution algorithm (DE) to optimize ELM for improving detection precision. In this paper, Autoencoder is used to reduce the dimensionality of the measurement data, which makes the low-dimensional data information basically and fully represent high-dimensional data. We verify the performance of the proposed method on IEEE bus systems, and prove that the proposed method can effectively detect such unobservable attack.