摘要
基于非侵入式电力负荷检测与分解技术近年来得到广泛推广.选取14个稳态指标作为负荷特征,建立基于支持向量机(SVM)的非侵入式负荷印记识别模型,利用多分类支持向量机(multi-class SVM)的成对分类算法,对负荷印记进行了识别,随机抽取数据进行测试,结果表明方法能够更准确地识别负荷印记,说明所提出的模型和方法具有较高的有效性和正确性.
Non-invasive power load detection and decomposition technology has been widely promoted in recent years.In this paper,14 steady-state indexes are selected as load characteristics,and a non-invasive load imprint recognition model based on support vector machine(SVM) is established.The load imprint is recognized by using the paired classification algorithm of multi-class support vector machine(multi-class SVM),and the data are randomly extracted for testing.The results show that the method can identify the load mark more accurately,which indicates that the model and method proposed in this paper have higher effectiveness and correctness.
作者
李玲娜
邵子娟
金垚灯
LI Ling-na;SHAO Zi-juan;JIN Yao-deng(School of Science,Southwest Petroleum University,Chengdu 610500,China)
出处
《数学的实践与认识》
北大核心
2019年第20期200-208,共9页
Mathematics in Practice and Theory
基金
教育部产学合作协同育人项目(201802151044)
关键词
负荷印记
模式识别
多分类支持向量机
成对分类算法
RBF核函数
load stamp
pattern recognition
multi-classification support vector machine
paired classification algorithm
rbf kernel function