数字电网风险数据多层级别挖掘模型构建  

Construction of Multi-level Mining Model for Digital Power Grid Risk Data

作　　者：刘明伟 任昊文 梁子健 刘竹青 LIU Ming-wei;REN Hao-wen;LIANG Zi-jian;LIU Zhu-qing(Platform Security Branch of CSG Digital Grid Research Institute Co.,Ltd.,Guangzhou 510000 China)

机构地区：[1]南方电网数字电网研究院有限公司平台安全分公司,广东广州510000

出　　处：《自动化技术与应用》2025年第3期96-99,共4页Techniques of Automation and Applications

基　　金：南网数研院平台安全分公司数据共享服务管理工具建设项目(JY-KF-03-GX-21-011-TQ)。

摘　　要：电网风险数据挖掘时由于其数据量大、数据类型多等特性,使得电网风险数据挖掘的及时性和安全性较差。为了提高数字电网的数据利用率,提出数字电网风险数据多层级别挖掘模型构建方法。通过平均比率法处理缺失数据,结合协同过滤思想完成数据缺失值填充。采用在线字典学习算法,消除噪声数据。在三角模糊数的基础上挖掘风险过程中影响因素之间的交互关系,划分出层级关系。建立数字电网数据多层阶梯风险挖掘模型,实现数字电网数据的挖掘。实验结果表明,所提方法的风险挖掘成本系数低于0.4、挖掘风险数据量为4800 bit且数据挖掘时间较短、挖掘的效率高。The timeliness and security of power grid risk data mining are poor because of its large amount of data and many types of data.In order to improve the data utilization of digital power grid,a multi-level mining model for digital power grid risk data is proposed.The missing data is processed by the average ratio method,and the missing data is filled with the idea of collaborative filtering.The online dictionary learning algorithm is adopted to eliminate noise data.On the basis of triangular fuzzy number,the interaction between influencing factors in the risk process is mined and the hierarchical relationship is divided.The multi-level ladder risk mining model of digital power grid data is established to realize data mining of digital power grid.The experimental results show that the cost coefficient of risk mining of the proposed method is lower than 0.4,the amount of risk data mining is 4800 bit,and the data mining time is short,and the mining efficiency is high.

关 键 词：数据挖掘 数字电网 缺失值填充 在线字典学习算法 数据风险挖掘模型 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]