摘要
信息系统传输故障、恶意攻击和物理系统故障都将威胁智能配电网的正常运行。现有研究未充分考虑信息系统与配电网融合造成的差异,特别是缺乏对信息系统故障与配电网物理故障进行统一辨识的方法。文中利用MATLAB和OPNET建立交互作用分析模型,对发生于信息物理系统中的故障进行定量分析;进而,提出基于多维尺度分析(multi-dimensional scaling,MDS)和局部异常因子(local outlier factor,LOF)的配电网故障辨识算法,实现信息系统故障和配电网物理故障的统一识别,以及对配电网故障区域和故障通信节点的定位。利用交互作用分析模型得到的仿真结果验证了对于智能配电网故障类型的分析,配电网物理故障和通信节点故障的辨识结果验证了本文算法的有效性。
An information transmission failure and malicious attack,as well as the physical system fault happening in a cyber physical system,will probably threaten the normal operation of smart distribution network.Unfortunately,little attention has been given to the interaction between cyber system and distribution network,especially the lack of an original method,of uniformly distinguishing between the cyber system failure and distribution network fault.An interaction model for the smart distribution network based on MATLAB and OPNET is developed to perform quantitative analysis of faults.A new fault identification algorithm of distribution network based on multi-dimensional scaling（MDS）and local outlier factor（LOF）algorithm is proposed.The simulation results of interaction analyzing model have validated the analysis of fault types in the smart distribution network.The results of case studies show that the proposed algorithm can not only implement the identification of cyber system failure and distribution network fault,but also locate the fault area of distribution network and the fault communication node.The simulation results show good consistency with theoretical analysis.
出处
《电力系统自动化》
EI
CSCD
北大核心
2016年第17期44-50,共7页
Automation of Electric Power Systems
基金
国家自然科学基金资助项目(61233008)
湖南省科技重大专项(2015GK1002)
湖湘青年英才项目(2015RS4022)~~
关键词
智能配电网
信息物理系统
多维尺度分析
局部异常因子检测
故障辨识
smart distribution network
cyber physical system
multi-dimensional scaling
local outlier factor
fault identification
