期刊文献+

基于决策树支持向量机的家用典型负荷分类 被引量:1

下载PDF
导出
摘要 家用典型负荷的特征分析和分类已成为智能用电领域的重点方向,从电路特性上解决家用典型负荷特征分析和分类方法问题,是基于多组家用电器实测波形,对电阻型、整流电子型和电机型三类典型家用负荷的电量波形特征进行分析,并提取功率因数、电流谐波总畸变率、50%电流幅值占比等参数构成组合负荷印记。利用决策树支持向量机(SVM)对家用典型负荷进行分类研究,并用300组典型负荷的实测负荷印记数据和单一功率因数数据对分类方法进行了验证,结果表明:所提负荷印记代表性强,分类方法准确率高达97.67%。先进行电器分类的判别方法,提高了对电器识别的准确性。 The characteristic analysis and classification of typical household loads has become a key direction in the field of intelligent electricity consumption.Solving the problem of characteristic analysis and classification of household typical load in terms of circuit characteristics is based on the measured waveforms of several groups of household appliances.The electrical waveform characteristics of three types of typical household loads,namely,resistance type,rectifier electronic type and motor type,are analyzed.The parameters such as power factor,total current harmonic distortion rate and 50%current amplitude ratio are extracted to form the combined load imprint.The typical household load is classified by decision tree support vector machine(SVM),and the classification method is verified by 300 groups of measured load imprint data and single power factor data.The results show that the proposed load imprint is highly representative and the accuracy of the classification method is as high as 97.67%.The discrimination method of electrical appliance classification is carried out to improve the accuracy of electrical appliance identification.
出处 《科技创新与应用》 2021年第34期24-27,共4页 Technology Innovation and Application
基金 国家创新训练计划(编号:201911319001)。
关键词 家用负荷 负荷印记 支持向量机 分类 household load load imprint support vector machine classification
  • 相关文献

参考文献7

二级参考文献38

共引文献58

同被引文献6

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部