Intrusion Detection of Integrated Energy System Based on Whitening PCA

• Published in: 2023 IEEE International Conference on Power Science and Technology (ICPST) pp. 745 - 750
• Main Authors: Xia, Yi, Ma, Hongxia, Yu, Xijuan, Wu, Yanhua, Yang, Xunfa, Xue, Liang, Zhu, Hong, Zhang, Peiqing
• Format: Conference Proceeding
• Language: English
• Published: IEEE 05.05.2023
• Subjects: Classification algorithms; Dimensionality reduction; extract the features; Feature extraction; integrated energy system; Intrusion detection; Network security; Support vector machines; the principle component value; Training; whitening PCA
• DOI: 10.1109/ICPST56889.2023.10165036

Due to the diversity of intelligent measuring equipment and the risk of subsystem communication, the integrated energy system is vulnerable to network attacks. Aiming at the security problem of integrated energy system, an intrusion detection method based on whitening PCA is proposed in this paper,based on the analysis of the structure of the integrated energy system and the characteristics of the integrated energy information physical system. Firstly, the dataset is preprocessed based on the normalization method, using whitening algorithm and PCA algorithm to extract the features and reduce the dimension of the data.The principal component value is obtained after data dimensionality reduction. Then support vector machine is used to classify and train the dimensionality reduction data set.The classification decision function is obtained for intrusion.Through the analysis of numerical example, the effectiveness is verified of the proposed method in intrusion detection.The detection rate is higher and the detection time is shorter.