计及非负和低秩特性的用电数据缺失值插补

打开文本图片集

中图分类号：U469.72+2 文献标志码：A 文章编号：1000-582X（2025）09-001-11

Abstract： With the widespread deployment smart meters，power grids have accumulated vast amounts raw electricityconsumption data. However，data loss remains a challenge due to the complex operational environments data acquisition equipment. This study addresses the problem incomplete electricity consumption data by accounting for the influence Gaussian noise and proposing a robust completion method.First，a electricity consumptiondata matrix is constructedby reorganizing thesequencesindividual users，and the idealelectricity data matrix is approximated using nonnegative matrix factorization （NMF）.Second，both the Frobenius norm and the nuclear normare employed to regularize the Gaussian noise and promote low-rank characteristics the ideal matrix，thereby formulating an optimization model.Finally， within a block coordinate descent framework，the EM algorithm and a direct updating method are applied alternately to update the matrix factors derived from NMF， enabling accurate and complete data reconstruction. Simulation and experimental results validate the proposed

algorithm’seffectivenessand accuracy.

Keywords： electricity consumption data; nonnegative matrix factorization; norm; block coordinate descent; matrix completion

近年来，随着数字化技术在智能电网中的广泛应用，电网公司在电力系统的运行中积累了大量数据，其中用户侧大数据占很大比重。（剩余12346字）