摘要
针对电网客服中心的网络安全问题,研究搭建了实时数据自动监测模型。首先,引入类内距离最小与类间距离最大的思想来对局部保持投影算法进行改进,用于检测异常数据。然后利用模糊规则和层次分析法来对电网实时数据自动监测进行优化。结果表明,改进局部保持投影算法的运行时间与传统局部保持投影算法相差不大。改进局部保持投影算法的检测准确率更高,为99.43%。基于层次分析的实时数据监测算法的监测准确率始终最高,在数据样本集为500时,准确率为99.84%,召回率为99.62%。实验结果证明了所提实时数据自动监测模型的异常数据检测和实时数据监测性能较好,有助于提高电网的安全性和稳定性。
A real-time data automatic monitoring model is developed to address the network security issues of the power grid customer service center.Firstly,the idea of minimizing intra class distance and maximizing inter class distance is introduced to improve the local preserving projection algorithm for detecting abnormal data.Then,fuzzy rules and Analytic Hierarchy Process are used to optimize the realtime data automatic monitoring of the power grid.The results indicate that the running time of the improved local preserving projection algorithm is not significantly different from that of the traditional local preserving projection algorithm.The improved local preserving projection algorithm has a higher detection accuracy of 99.43%.The real time data monitoring algorithm based on Analytic Hierarchy Process always has the highest monitoring accuracy,with an accuracy of 99.84%and a recall rate of 99.62%when the data sample set is 500.The experimental results demonstrate the abnormal data detection and real-time data monitoring performance of the proposed real-time data automatic monitoring model are good,which helps to improve the safety and stability of the power grid.
作者
侯昝宇
HOU Zanyu(State Grid Liaoning Electric Power Supply Co.,Ltd.,Shenyang 110006,Liaoning,China)
出处
《自动化技术与应用》
2026年第2期138-142,共5页
Techniques of Automation and Applications
基金
辽宁省教育厅面上项目(LJKZ0138)。
关键词
流形学习
数据自动监测
电网客服
层次分析法
局部保持投影
manifold learning
automatic data monitoring
power grid customer service
analytic hierarchy process
locally preserving projection
