摘要
传统电网输电量异常数据检测方法存在检测准确率较低的缺陷,为了解决上述问题,提出基于并行分类算法的电网输电量异常数据检测方法。以现有电网输电量异常数据为依据,基于信息熵提取输电量异常数据特征,采用无线mesh网络结构采集电网输电量数据,通过并行分类算法预处理输电量数据,得到输电量数据分类结果,经过随机森林模型检测并修复输电量异常数据,实现基于并行分类算法的电网输电量异常数据检测。仿真对比实验结果表明,提出的电网输电量异常数据检测方法极大地提升了检测准确率,充分说明所提方法具备更好的检测效果。
In order to solve the above problems,a new method based on parallel classification algorithm is proposed.Based on the existing abnormal data of power transmission,the characteristics of abnormal data of power transmission are extracted based on information entropy,and the data of power transmission is collected by wireless mesh network structure.The data of power transmission is preprocessed by parallel classification algorithm,and the result of data classification is obtained.The abnormal data of power transmission is detected and repaired by the forest model,and the power transmission based on parallel classification algorithm is realized abnormal power data detection.The simulation results show that the proposed method greatly improves the detection accuracy,which fully shows that the proposed method has better detection effect.
作者
熊学锋
张涵
荣功立
宋凯
孔德诗
XIONG Xuefeng;ZHANG Han;RONG Gongli;SONG Kai;KONG Deshi(State Grid Sichuan Electric Power Corporation Power Supply Service Center,Chengdu 610041,China)
出处
《电子设计工程》
2020年第24期91-94,99,共5页
Electronic Design Engineering
基金
国家重点研究计划项目(2017YFA0500301)
四川省自然科学基金项目(SGSCKE00JYJS1600059)。
关键词
并行分类算法
电网
输电量异常
异常数据
检测
parallel classification algorithm
power grid
transmission difference
abnormal data
detection
