摘要
为有效监测电力负荷异常变化情况,设计一种基于配电网负荷潮流约束的用电异常反向监测方法。采集配电网中的用户多项数据并完成归一化处理,采用K-means聚类算法完成用电数据聚类,过程中使用Redis集群维护用电数据流,设计潮流约束计算流程,修正电压,完成配电网中负荷功率倒推。利用BP神经网络设计内置激活函数,输入多项采集数据后,输出用电是否异常的单一监测结果。结果表明:设计方法对于用电有无异常行为具有较高的分类精准度,较传统的监测方法异常检出比例提高了24.8%,设计方法P-R曲线的AP值比传统方法高14.3%。
In order to effectively monitor the abnormal change of power load,a reverse monitoring method of abnor⁃mal power consumption based on the load power flow constraint of distribution network was designed.A number of user data in the distribution network were collected and normalized.The K-Means clustering algorithm was used to cluster the power consumption data.In the process,Redis cluster was used to maintain the power consumption data stream,design the calculation process of power flow constraint,correct the voltage,and complete the load power re⁃versal in the distribution network.The built-in activation function was designed by using the BP neural network,and after inputting multiple collected data,a single monitoring result of whether the electricity consumption was ab⁃normal was output.The experimental results showed that the design method had a high classification accuracy for the abnormal behavior of electricity consumption,and the abnormal detection ratio was increased by 24.8%com⁃pared with the traditional monitoring method,and the AP value of the P-R curve of the design method was 14.3%higher than that of the traditional method.
作者
刘栋果
LIU Dongguo(State Grid Sichuan Electric Power Corporation Co.,Ltd.,Chengdu 610041,China)
出处
《粘接》
CAS
2024年第6期170-173,共4页
Adhesion
关键词
负荷特性
潮流约束
反向监测
BP神经网络
负荷倒推
load characteristics
current constraints
reverse monitoring
bp neural network
load reversal
