Title :
Intrusion Detection System for Electric Power Information Network Based on Improved Ball Vector Machine
Author :
Wang Yufei ; Zhou Liang ; Wang Jing
Author_Institution :
Dept. of Inf. & Commun., China Electr. Power Res. Inst., Beijing, China
Abstract :
It is helpful to enhance the information security of Electric Power Information Network (EPIN) that researching the intrusion detection technology. In order to achieve efficient intrusion detection for EPIN, an Intrusion Detection System (IDS) based on the improved Ball Vector Machine (BVM) is proposed. In this paper, the IDS and its detection rules are automatically generated by the way that the improved BVM is used to train the historical data. In the IDS based on the improved BVM, the BVM is used to reduced time-consuming, in addition, in order to enhance the intrusion detection accuracy, the Particle Swarm Optimization (PSO) is used to search the best training parameters of BVM in training process. Finally the experiment based on EPIN data shows that the IDS based on the improved BVM has better performance than the traditions.
Keywords :
particle swarm optimisation; power engineering computing; power system security; security of data; support vector machines; BVM; EPIN data; IDS; PSO; electric power information network; historical data training process; improved ball vector machine; information security; intrusion detection system; particle swarm optimization; support vector machine; Accuracy; Intrusion detection; Optimization; Power systems; Real-time systems; Support vector machines; Training; controlled islanding; power flow tracing; power system; splitting boundary;
Conference_Titel :
Intelligent Systems Design and Engineering Applications, 2013 Fourth International Conference on
Conference_Location :
Zhangjiajie
Print_ISBN :
978-1-4799-2791-3
DOI :
10.1109/ISDEA.2013.488
