基于深度神经网络的配电自动化网络安全态势辨识  

Security Situation Identification of Distribution Automation Network Based on Deep Neural Network

作　　者：江灏 张绮华 宋晓阳 孙冉冉 JIANG Hao;ZHANG Qihua;SONG Xiaoyang;SUN Ranran(Ningbo Power Supply Company of State Grid Zhejiang Electric Power Co.,Ltd.,Ningbo 315012,China)

机构地区：[1]国网浙江省电力有限公司宁波供电公司,浙江宁波315012

出　　处：《微型电脑应用》2024年第7期68-71,共4页Microcomputer Applications

基　　金：国网浙江宁波供电公司科技项目(KJCXX20210401)。

摘　　要：在进行配电网络安全态势自动辨识时,现有方法主要通过单隐藏层神经网络提取数据特征,使得辨识结果标准误差较大。为此,提出一种基于深度神经网络的配电自动化网络安全态势自动辨识算法。从静态安全性、动态安全性两方面入手建立完整的评价指标体系,描述电网运行状态。结合层次分析法和改进熵权法计算网络安全态势评估值,合理划分风险等级。使用深度神经网络构建辨识模型,提取多层级网络数据的深层次特征,得到网络安全态势辨识结果。根据引力函数和适应度函数,搜索最优安全态势辨识结果。实验结果表明,所提算法的安全态势自动辨识结果与BP神经网络辨识算法、RBP神经网络辨识算法相比,标准误差降低了27个百分点、29个百分点,安全态势辨识准确性可达99.55%。In the automatic identification of distribution network security situation,the existing methods mainly extract data features through a single hidden layer neural network,which makes the standard error of identification results larger.Therefore,this study proposes an automatic identification algorithm of distribution automation network security situation based on deep neural network.From static security and dynamic security,a complete evaluation index system is established to describe the operation state of the power grid.The evaluation value of network security situation is calculated by combining the analytic hierarchy process and the improved entropy weight method,and the risk level is reasonably divided.A deep neural network is used to build an identification model,extract the deep level characteristics of multi-level network data,and obtain the network security situation identification results.The gravity function and fitness function are used to search the optimal security situation identification results.The experimental results show that compared with BP neural network identification algorithm and RBP neural network identification algorithm,the standard error of the automatic situation identification algorithm is reduced by 27 percentage points,29 percentage points,and the accuracy of security situation identification is up to 99.55%.

关 键 词：深度神经网络 配电网 网络安全 态势感知 自动辨识 评价指标 

分 类 号：TM93[电气工程—电力电子与电力传动]