张量分解理论下的10kV配电网电压数据修复方法  

Voltage Data Repair Method for 10kV Distribution Network under Tensor Decomposition Theory

作　　者：王晓虎 WANG Xiaohu(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院电气学院,上海201306

出　　处：《电力系统及其自动化学报》2022年第7期148-152,共5页Proceedings of the CSU-EPSA

摘　　要：完整准确的运行数据是充分利用智能电网大数据的基础。针对当前配电网由于传感器不稳定等问题造成数据缺失问题,研究了一种基于张量分解理论的10 kV配电网电压数据修复方法。该方法首先提出了利用线路拓扑结构、时序数据和相似日数据构建时空张量的策略,充分挖掘现有数据中的内在关系,针对张量缺失使用规则化均值的方法进行初始化,利用贝叶斯CP分解张量并恢复缺失数据,以完成数据修复的目的,进而提高配电网电压数据的质量。算例结果表明:所提方法可以有效修复缺失数据,数据误差在0.5%以下。The complete and accurate operation data is the basis for fully utilizing the smart grid big data.To solve the problem of missing data in distribution network at present due to factors such as sensor instability,a tensor decomposi⁃tion theory-based voltage data repair method for 10 kV distribution network is studied.First,a strategy for constructing a spatio-temporal tensor is proposed using the line topology,timing data and similar day data,the inherent relationship within the existing data is fully mined,and the missing tensor is initialized using a regularized averaging method.The tensor is decomposed using the Bayesian CP decomposition,and the missing data is restored to realize the objective of data repair,thereby improving the quality of distribution network voltage data.The results of an example show that the proposed method can effectively repair the missing data,with a data error of lower than 0.5%.

关 键 词：数据修复 规则化均值 时空张量 智能电网大数据 

分 类 号：TM73[电气工程—电力系统及自动化]