基于多节点协同的电力网络靶场混合入侵识别方法
    点此下载全文
引用本文:郭舒扬,王宁,覃岩岩.基于多节点协同的电力网络靶场混合入侵识别方法[J].计算技术与自动化,2024,(1):154-159
摘要点击次数: 43
全文下载次数: 0
作者单位
郭舒扬,王宁,覃岩岩 (海南电网有限责任公司 信息通信分公司海南 海口 570203) 
中文摘要:研究了基于多节点协同的电力网络靶场混合入侵识别方法,提升电力网络对混合入侵行为的防御能力。依据多节点协同的分层协作结构,设置电力网络靶场训练平台内的通信节点作为感知节点,利用感知节点感知混合入侵数据,将感知结果传送至电力网络靶场训练平台的中心节点,利用中心节点融合混合入侵感知数据形成聚合节点。协同融合层的聚合节点将协同融合结果传送至识别层,识别层利用混合入侵识别模块,依据K-means聚类算法对混合入侵数据的聚类结果,构建C4.5决策树,利用决策树输出电力网络靶场混合入侵识别结果。实验结果表明,该方法可以精准识别电力网络靶场混合入侵行为,识别精度高于98%。
中文关键词:多节点协同  电力网络靶场  混合入侵  识别方法  K-means聚类  C4.5决策树
 
Hybrid Intrusion Identification Method for Power Network Range Based on Multi-node Cooperation
Abstract:A hybrid intrusion identification method based on multi-node cooperation in power network shooting range is studied to improve the defense capability of power network against hybrid intrusion behavior. Based on hierarchical collaborative structure of multi-node cooperative, set up the electric power network as perception communication node within the range training platform, using the data from the mixed aware of node invasion will perceive the results sent to the center of the power network range training platform node, using data fusion hybrid intrusion awareness center node form aggregation node. The aggregation node of the collaborative fusion layer sends the collaborative fusion results to the recognition layer, which uses the hybrid intrusion recognition module to cluster the hybrid intrusion data according to the K-means clustering algorithm, constructs the C4.5 decision tree, and uses the decision tree to output the hybrid intrusion recognition results of the power network range. The experimental results show that the method can accurately identify the hybrid intrusion behavior of power network range, and the recognition accuracy is higher than 98%.
keywords:multi-node collaboration  power network range  hybrid intrusion  identification method  K-means clustering  C4.5 decision tree
查看全文   查看/发表评论   下载pdf阅读器