基于信息差异图模型的电力调度自动化系统组件故障溯源方法  

Component Fault Tracing of Power Dispatching Automation System Based on Information Difference Graph Model

作　　者：高欣[1] 任昺 张浩 刘蒙 李军良 徐建航 GAO Xin;REN Bing;ZHANG Hao;LIU Meng;LI Junliang;XU Jianhang(School of Artificial Intelligence,Beijing University of Posts and Telecommunications,Haidian District,Beijing 100876,China;State Grid Jibei Electric Power Company Limited,Xicheng District,Beijing 100054,China;NARI Group(State Grid Electric Power Research Institute)Corporation,Haidian District,Beijing 100192,China)

机构地区：[1]北京邮电大学人工智能学院,北京市海淀区100876 [2]国网冀北电力有限公司,北京市西城区100054 [3]南瑞集团(国网电力科学研究院)有限公司,北京市海淀区100192

出　　处：《电网技术》2021年第12期4808-4817,共10页Power System Technology

基　　金：国家电网有限公司总部科技项目(5700-2020040366A-0-0-00)。

摘　　要：电力调度自动化系统作为复杂信息物理系统,规模庞大且逻辑拓扑难以获得,一旦其组件发生故障,系统将出现雪崩式告警,现有典型故障溯源方法严重依赖于专家经验和逻辑拓扑关系,故障溯源困难。提出一种基于信息差异图模型的电力调度自动化系统故障溯源方法。首先,通过k-means算法获得所采集每个数据特征时间序列的聚类中心,将其作为划分离散区间的端点,计算区间均值作为离散化结果。然后,根据告警前后的信息度量变化率构建信息差异矩阵,该矩阵元素分别由系统组件自身信息熵和组件间传递熵的归一化差异比率组成。最后,根据告警信息变化较高的特征及特征间的交互信息建立信息差异图模型,定义图模型中节点状态及链接关系,构建节点及其权值和故障程度之间的映射表,标记表中故障程度排名靠前的组件并回溯到图模型中组件间交互关系进行故障溯源。在仿真数据、某开源分布式系统数据集和某电网调度系统实采数据上的实验结果表明,与典型相关方法相比,所提方法在准确率、召回率等指标上具有显著优势。As a complex information physical system,the power dispatching automation system is large in scale and difficult to obtain its logical topology.Once a system component fault occurs,the system will appear an avalanche alarm,making the fault tracing difficult.To solve the problem that the typical fault tracing methods rely heavily on the expert experience and the logical topological relations,this paper proposes a fault tracing method of power dispatching automation system based on the information difference graph model.First,the clustering center of each data feature time series collected is obtained through the K-means algorithm,which is taken as the endpoint dividing the discrete intervals and the interval mean is calculated as the discretization result.Second,the information difference matrix is established according to the information measurement rate before and after the alarm.The matrix elements are composed of the information entropy of the system components and the normalized difference ratio of the transfer entropy between the components respectively.Finally,an information difference graph model is established according to the characteristics with high variation alarm information and the interaction information among them.The state and link relationship of the nodes in the graph model are defined;the mapping table between the nodes with their weights and the fault degree is constructed;the components in the table with the highest fault degree are marked;and the interaction relationship between components in the graph model is traced for the fault source.Experimental results on the simulation dataset,an open source distributed system dataset and actual data of a power grid dispatching system show that this method is feasible and effective.

关 键 词：电力调度自动化系统 组件故障溯源 信息度量 信息差异图模型 

分 类 号：TM721[电气工程—电力系统及自动化]