摘要
提出了1种基于模糊推理的物理与信息系统异常融合方法,以检测智能电网的攻击事件。根据电力系统的物理约束,检测电力系统量测量中的异常数据,并计算物理系统异常度;对信息网络的通信数据进行监测,发现异常通信报文,计算信息系统异常度;根据智能电网中电力系统和信息网络的对应关系将物理和信息异常度进行关联匹配,采用模糊推理方法,融合计算系统的物理-信息异常度,检测攻击事件。仿真实验中对IEEE 14、39和118节点系统模拟不同强度的错误数据注入攻击,实验结果表明通过对电力系统和信息系统异常检测结果进行模糊推理,可以保证在较低漏报率条件下有效降低误报率。
A cyber-physical fuzzy inference method （CPFI） is proposed to detect the attack in smart grid. The measurements of power system are collected to detect the false data and evaluate the physical abnormality according to the electrical laws. The network traffic is monitored to detect the anomaly packets and evaluate cyber abnormality. Then, the cyber and physical abnormalities are associated into pairs by utilizing the relevance of power grid and information network in smart grid. The fuzzy inference algorithm is designed to fuse the cyber-physical abnormality pairs and figure out attack behaviors. Several false data injection attacks are simulated in IEEE 14, 39 and 118 bus systems. It is demonstrated that the CPFI outperforms current abnormal detection methods in power grid and information network with much lower false negative rate.
出处
《中国科技论文》
CAS
北大核心
2016年第14期1619-1625,共7页
China Sciencepaper
基金
国家自然科学基金资助项目(61203174)
高等学校博士学科点专项科研基金资助项目(20110201120010)
