摘要
目前电器种类迅速增多,各类电器的负荷特征相似度随之提高,难以准确辨识。针对这一现状,提出了一种基于复合特征的非侵入式电力负荷识别方法。该方法对传统单目标函数进行了改进优化,并且基于电器功率和正交电流谐波特征,进一步建立了基于遗传算法的复合特征目标函数模型,通过遗传迭代多目标寻优获得最优个体,实现精确的负荷识别。实验结果表明,所提出的基于复合特征的非侵入式电力负荷识别方法识别准确率较高,可以满足实际应用需求。
The power load features of various electrical equipment become increasingly similar to each other,along with the rapid in-crease of category numbers.Consequently,it becomes more difficult to implement accurate load identification.With regard to this condi-tion,a non-invasive power load monitoring method(NILM)based on composite electrical features is presented,with improvement com-pared to the conventional single-object function.Based on the load features of both electrical power and current harmonics,the compos-ite-objective function model is established in combination with the genetic algorithm.Then the genetic iterative multi-objective optimiza-tion is utilized to obtain the optimal results and realize accurate load identification.The experimental results demonstrate that the pro-posed non-invasive power load identification method based on composite features achieves high recognition accuracy and can meet the demands of practical applications.
作者
王传君
缪巍巍
曾锃
李世豪
蒋姝
WANG Chuanjun;MIAO Weiwei;ZENG Zeng;LI Shihao;JIANG Shu(Information and Telecommunication Branch,State Grid Jiangsu Electric Power Company,Nanjing Jiangsu 210024,China;School of Information and Communication Engineering,Nanjing Institute of Technology,Nanjing Jiangsu 211167,China)
出处
《电子器件》
CAS
北大核心
2023年第1期218-222,共5页
Chinese Journal of Electron Devices
关键词
负荷识别
复合特征
功率特征
电流谐波特征
遗传算法
load identification
composite features
power feature
current harmonics
genetic algorithm
