基于改进自注意力机制生成对抗网络的智能电网GPS欺骗攻击防御方法  

Defense method of smart grid GPS spoofing attack based on improved self-attention generative adversarial network

作　　者：李元诚[1] 杨珊珊 LI Yuancheng;YANG Shanshan(School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华北电力大学控制与计算机工程学院,北京102206

出　　处：《电力自动化设备》2021年第11期100-106,共7页Electric Power Automation Equipment

基　　金：中央高校基本科研业务费专项资金资助项目(2020YJ003)。

摘　　要：为了避免全球定位系统欺骗攻击(GSA)对相量测量装置造成的危害,提出了一种基于改进自注意力机制生成对抗网络(SAGAN)的智能电网GSA防御方法。首先,通过引入深度学习参数,构建了改进网络-物理模型,利用历史数据计算得到当前时刻的量测值。然后,在SAGAN的生成器和判别器网络中分别融入一个时间注意力模块,提出了一种用于实现网络-物理模型的改进SAGAN防御方法。通过训练改进SAGAN,得到一对判别器和生成器,利用判别器检测采集的量测值是否遭受GSA,当检测到攻击时,利用生成器生成的数据替换欺骗数据,从而实现智能电网对GSA的主动防御。最后,基于IEEE 14节点和IEEE 118节点系统进行仿真测试,结果验证了所提方法的可行性和有效性。In order to avoid the damage caused by GSA(Global positioning system Spoofing Attack)on PMU(Phasor Measurement Unit),a defense method of smart grid GSA based on improved SAGAN(Self-Attention Generative Adversarial Network)is proposed.Firstly,by introducing deep learning parameters,the improved cyber-physical model is built,and the measurement value at the current moment is calculated by using historical data.Then,a time attention module is incorporated into SAGAN’s generator and discriminator respectively,and an improved SAGAN defense method is proposed to implement the cyber-physical model.By training the improved SAGAN,a pair of discriminator and generator is obtained.The discriminator is used to detect whether the collected measurement values are subject to GSA.When the attack is detected,the data generated by the generator is used to replace the spoofing data,so as to realize the active defense of smart grid GSA.Finally,simulation tests are carried out based on IEEE 14-bus system and IEEE 118-bus system,and the results verify the feasibility and effectiveness of the proposed method.

关 键 词：智能电网 全球定位系统 欺骗攻击 相量测量装置 生成对抗网络 防御方法 

分 类 号：TM761[电气工程—电力系统及自动化]