基于神经网络架构的电力系统DoS与网络攻击检测和隔离  

Power System DoS and Network Attack Detection and Isolation Based on Neural Network Architecture

作　　者：陈媛媛 汪伟鸣 温安平 吴俊 李映 CHEN Yuanyuan;WANG Weiming;WEN Anping;WU Jun;LI Ying(State Grid XuanCheng Electric Power Supply Company,Xuancheng Anhui 242000,China)

机构地区：[1]国网宣城供电公司,安徽宣城242000

出　　处：《电子器件》2024年第5期1350-1354,共5页Chinese Journal of Electron Devices

摘　　要：针对电力系统拒绝服务与网络攻击的检测和隔离进行了分析,分析了电力系统的DoS检测和隔离动态控制模型,该模型适用于常规的动态网络。并提出了网络攻击模型,说明了网络攻击检测和隔离的方法,并针对电力系统的具体情况提出了网络攻击检测架构,说明了利用神经网络等算法实现电力系统数据预测的模型和供给检测的流程。以自动发电控制AGC为案例,说明电力系统网络攻击检测隔离方法的有效性。Aiming at the detection and isolation of denial of service and network attacks in power systems,the dynamic control model of DoS detection and isolation in power systems is analyzed,which is suitable for conventional dynamic networks.A network attack model is proposed,the method of network attack detection and isolation is explained,a network attack detection architecture is proposed according to the specific situation of the power system,and the model and supply detection process of using neural network and other algorithms to realize power system data prediction are explained.Taking the automatic power generation control AGC as an example,the effectiveness of the detection and isolation method of power system network attack is illustrated.

关 键 词：神经网络架构 DOS攻击 攻击检测和隔离 自动发电控制 

分 类 号：TM73[电气工程—电力系统及自动化] TK01[动力工程及工程热物理]