湖南电力 ›› 2024, Vol. 44 ›› Issue (5): 68-73.doi: 10.3969/j.issn.1008-0198.2024.05.011

• 新技术及应用 • 上一篇    下一篇

基于业务基线的电力物联终端智能辨识方法

李明光1, 杨芳僚2, 李浩志2, 黄鑫2, 杨圣洪1   

  1. 1.湖南大学信息科学与工程学院,湖南 长沙 410082;
    2.国网湖南省电力有限公司信息通信分公司,湖南 长沙 410004
  • 收稿日期:2024-05-30 修回日期:2024-07-02 出版日期:2024-10-25 发布日期:2024-11-06
  • 通信作者: 李明光(1999),女,主要从事网络空间安全研究工作。
  • 基金资助:
    国家重点研发计划(2021YFF0901001); 国网湖南省电力有限公司重大科技项目(5216A622000G)

Intelligent Identification Method for Power Internet of Things Terminals Based on Business Baseline

LI Mingguang1, YANG Fangliao2, LI Haozhi2, HUANG Xin2, YANG Shenghong1   

  1. 1. College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, China;
    2. State Grid Hunan Electric Power Company Limited Information and Communication Company, Changsha 410004, China
  • Received:2024-05-30 Revised:2024-07-02 Online:2024-10-25 Published:2024-11-06

摘要: 针对新型电力系统数字化建设的关键环节,提出一种基于业务基线的电力物联终端智能辨识方法,采集电力物联终端流量报文并分析数据分布特征,定义电力物联终端业务基线;对须识别的终端设备提取特征向量,与业务基线进行匹配,从而实现对使用Modbus RTU协议进行通信的电力物联终端设备的智能辨识。为验证方法有效性,采集10种电力物联网终端设备在单位时间内连续10次接收自主构建的Modbus消息帧,并返回相应的应答数据。结果表明,当阈值高于0.08时,识别精度较高,所提方法无需繁杂的特征提取,且识别精度较高。

关键词: 电力物联网, 终端设备, 智能辨识, 业务基线, 特征向量

Abstract: Aiming at the key link of the digital construction of the new power system, a smart identification method for power internet of things(IoT)terminals based on business baselines is proposed. The method collects flow messages from power IoT terminals, analyzes data distribution characteristics, and defines the business baseline of power IoT terminals. The feature vectors from the terminal devices that need to be identified are extracted, matching them with the business baseline, so as to achieve intelligent identification of power IoT terminal devices that communicate using Modbus RTU protocol. To verify the effectiveness of the method, 10 types of power internet of things terminal devices are collected to receive autonomously constructed Modbus message frames continuously 10 times per unit time, and corresponding response data are returned. The results show that when the threshold is higher than 0.08, the recognition accuracy is high, and the proposed method does not require complex feature extraction, and the recognition accuracy is high.

Key words: electric power internet of things, terminal devices, intelligent recognition, business baseline, feature vectors

中图分类号: