计算机应用与软件2018，Vol.35Issue(2)：54-59,6.DOI:10.3969/j.issn.1000-386x.2018.02.009
基于L0稀疏超图半监督学习的异常用电行为识别
ABNORMAL ELECTRICITY POWER CONSUMPTION RECOGNITION BASED ON L0 SPARSE HYPERGRAPH SEMI-SUPERVISED LEARNING
摘要
Abstract
It is a very important problem for the electric power department to carry out the automatic analysis of theabnormal power consumption behavior .This paper proposed a semi-supervised classification algorithm based on L0 sparsehypergraph learning.Different from the pair-wise link in traditional graph model, hyperedge in hypergraph is a subset ofdata points sharing with same attributes, which is beneficial for representing the complex correlations between the powerload profiles of different users.In practices, the size of hyperedges constructed by common K-nearest-neighbor method issame for all vertexes, which make it hard to capture the non-uniform distributed characteristics of the power load dataset .In order to solve this problem, this thesis established a hypergraph model based on L0 sparse reconstruction.The modelcreated a hyperedge for each user's data.It was more advantageous to match the distribution of user data by adaptivelyselecting multiple samples closely related to the current user through sparse decomposition of L0 norm constraints.Then,a semi-supervised classification model of the hypergraph Laplacian regular constraint was constructed , and a smallamount of calibration sample data was used to determine whether the user's electrical behavior was abnormal.We selected the actual electricity consumption data of more than 300 days in a city of more than 8 900 residents as the testsample set.Experimental results verified the effectiveness of the proposed method.关键词
用电行为分析/L0稀疏/超图/半监督学习Key words
Electricity behavior analysis/L0 sparsity/Hypergraph/Semi-supervised learning分类
信息技术与安全科学引用本文复制引用
郭志民,袁少光,孙玉宝..基于L0稀疏超图半监督学习的异常用电行为识别[J].计算机应用与软件,2018,35(2):54-59,6.基金项目
国家自然科学基金项目(61300162) （61300162）
国家电网公司2016年科技项目. （）
