电子学报 ›› 2020, Vol. 48 ›› Issue (8): 1538-1544.DOI: 10.3969/j.issn.0372-2112.2020.08.011

• 学术论文 • 上一篇    下一篇

基于深度学习的配电网无线通信入侵检测系统

刘文军1, 郭志民2, 吴春明3, 阮伟4, 周伯阳2, 周宁2, 吕卓2   

  1. 1. 国网河南省电力公司, 河南郑州 450000;
    2. 国网河南省电力公司电力科学研究院, 河南郑州 450000;
    3. 浙江大学计算机科学与技术学院, 浙江杭州 310027;
    4. 浙江大学控制科学与工程学院, 浙江杭州 310027
  • 收稿日期:2018-04-18 修回日期:2018-06-06 出版日期:2020-08-25 发布日期:2020-08-25
  • 通讯作者: 阮伟
  • 作者简介:刘文军 男,1967年出生,华北电力大学通信工程专业毕业,高级工程师,国网河南省电力公司科信部通信处处长,从事电力通信工作29年.郭志民 男,1977年出生,本科,教授级高级工程师,国网河南省电力公司电力科学研究院设备状态评价中心副主任,研究方向为电力系统自动化、电力信息安全.

A Deep Learning Based Intrusion Detection System for Electric Distribution Grids

LIU Wen-jun1, GUO Zhi-min2, WU Chun-ming3, RUAN Wei4, ZHOU Bo-yang2, ZHOU Ning2, Lü Zhuo2   

  1. 1. State Grid Henan Electric Power Company, Zhengzhou, Henan 450000, China;
    2. State Grid Henan Electric Power Research Institute, State Grid Henan Electric Power Company, Zhengzhou, Henan 450000, China;
    3. College of Computer Science and Technology, Zhejiang University, Hangzhou, Zhejiang 310027, China;
    4. College of Control Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China
  • Received:2018-04-18 Revised:2018-06-06 Online:2020-08-25 Published:2020-08-25

摘要: 在采用无线通信接入的配电网中,入侵检测系统(IDS)通过分析通信网中传输数据来判断入侵事件.为提高检测的准确性,本文将深度学习理论应用于IDS,提出了一种面向配电网无线通信网络新型入侵检测系统,由带有门控循环单元、多层感知器和Softmax的循环神经网络组成.攻击测试基准实验结果表明IDS防御的有效性,在KDD99测试数据集上,其误报率为0.06%,总检出率为96.43%;在NSL-KDD测试数据集上,其误报率低至0.86%,总检出率则为99.33%.

关键词: 配电网, 无线网, 入侵检测, 深度学习, 递归神经网络

Abstract: In an electric power distribution grid using wireless communication access,IDS is used to decide system the intrusive event through analyzing the network transmission data.In this paper,to improve the detection accuracy,a deep learning theory is studied for the IDS in the wireless communication network of a power distribution grid.The proposed Recurrent Neural Network(RNN)model is composed of Gated Recurrent Unit(GRU),Multi-Layer Perceptron(MLP)and Softmax.The experimental results on the attack testing baseline demonstrate the effectiveness of the IDS defenses.In the KDD99 test data,its negative error rate and accuracy are with 0.06% and 96.43%,and in the NSL-KDD test data,those statistics are 0.86% with 99.33%,respectively.

Key words: electric distribution network, wireless network, intrusion detection, deep learning, recurrent neural network(RNN)

中图分类号: