«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

期数:

2025年05期

页码:

145-151

栏目:

人工智能

出版日期:

2025-05-10

文章信息/Info

Title:

Power IoT Equipment Fault Prediction Based on GraphSAGE Algorithm

文章编号:

1673-629X(2025)05-0145-07

作者:

李世豪1*; 曾锃1; 缪巍巍1; 夏元轶1; 刘鹏飞2; 赵海涛2

1. 国网江苏省电力有限公司信息通信分公司,江苏 南京 210024;

2. 南京邮电大学 通信与信息工程学院,江苏 南京 210003

Author(s):

LI Shi-hao1*; ZENG Zeng1; MIAO Wei-wei1; XIA Yuan-yi1; LIU Peng-fei2; ZHAO Hai-tao2

1. Information and Communication Branch,State Grid Jiangsu Electric Power Co. ,Ltd. ,Nanjing 210024,China;

2. School of Communications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China

关键词:

电力系统; 电力物联网; GraphSAGE算法; 电力物联设备故障; 有效预测

Keywords:

power system; power Internet of Things; Graph Sample and Aggregate algorithm; power IoT equipment fault; effective predic-tion

分类号:

TP277;TM73

DOI:

10.20165/j.cnki.ISSN1673-629X.2024.0397

摘要:

电力系统的安全稳定运行是保障国家能源安全和经济发展的关键,而这在很大程度上依赖于对电力物联设备故障的准确预测。 当前,随着电力物联网技术的发展,大量的数据被采集,但这些数据的潜在价值尚未得到充分挖掘,这在一定程度上限制了故障预测的准确性,影响了电力系统的可靠运行。 针对这一问题,该文提出了一种创新的基于GraphSAGE(Graph Sample and Aggregate)算法的电力物联设备故障预测。 该方法通过 PowerGraph 数据集,将电力物联设备故障场景细分为四类,利用 GraphSAGE 模型的特性,深入学习和分析节点特征与边特征,从而实现对物联设备故障的有效预测。 实验结果表明,该方法准确率达到 97. 5% ,相较于其它传统方法,准确率提高了 0. 39% ~ 6. 21% ,同时GraphSAGE 模型实现了快速训练。 该方法为电力物联设备安全稳定运行提供重要决策支持,能够对动态和相互联系的复杂系统进行更精细的分析,并增强电力系统运营部门对潜在干扰的预见和应对能力。

Abstract:

The safe and stable operation of power systems is crucial for ensuring national energy security and economic development. This stability largely depends on the accurate prediction of power IoT equipment faults. With the advancement of power Internet of Things (IoT) technology,a vast amount of data is being collected. However,the potential value of these data has not been fully tapped,limiting the accuracy of fault predictions and affecting the reliable operation of power systems. To address this issue,we propose an innovative method for power IoT equipment fault prediction based on the GraphSAGE ( Graph Sample and Aggregate) algorithm. Utilizing the PowerGraph dataset,the proposed method categorizes power IoT equipment fault scenarios into four types. It leverages the characteristics of the GraphSAGE model to deeply learn and analyze node and edge features,thereby effectively predicting power IoT equipment faults.Experimental results show that the proposed method achieves an accuracy rate of 97. 5% ,which is an improvement of 0. 39% to 6. 21% over other traditional methods. Furthermore,the GraphSAGE model enables rapid training,providing critical decision support for the safe and stable operation of power equipment. The proposed method allows for more refined analysis of dynamic and interconnected complex systems,enhancing the ability of power system operators to anticipate and respond to potential disturbances.

常用功能

更新日期/Last Update: 2025-05-10