基于角度分段线性近似和改进密度峰值聚类的户变关系识别  

Meter-to-transformer relationship identification based on APLR and ICFSFDP

作　　者：赵耀 付皖皖 陈冉 张涛 ZHAO Yao;FU Wanwan;CHEN Ran;ZHANG Tao(College of Electrical Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Electric Power Research Institute,State Grid Shanghai Municipal Electric Power Company,Shanghai 200437,China)

机构地区：[1]上海电力大学电气工程学院,上海200090 [2]国网上海市电力公司电力科学研究院,上海200437

出　　处：《电力科学与技术学报》2025年第1期113-125,共13页Journal of Electric Power Science And Technology

基　　金：国家自然科学基金(52377111);国家电网有限公司科技项目(B3094023000D);上海市自然科学基金(21ZR1425400)。

摘　　要：为解决因排查效率低、数据更新不及时等因素导致低压配电网户变关系连接形式与实际不符的问题,提出一种基于角度分段线性近似(anglepiecewiselinearrepresentation,APLR)和改进密度峰值聚类(improved clustering by fast search find of density peaks,ICFSFDP)相结合的户变关系识别方法。首先,根据电压曲线中相邻线段的角度变化量提取曲线的转折点,利用APLR对曲线进行自适应降维重构;随后,使用ICFSFDP算法对降维数据组展开聚类分析,在决策图中由拟合函数与坐标轴围成面积的最小值得到最优类簇数目,进而得到聚类和非聚类中心用户;最后,使用动态时间弯曲(dynamic time warping,DTW)距离计算聚类和非聚类中心用户之间的距离相似度,进而得到户变关系。将所提方法应用于模拟和真实数据中,均可证实所提方法的有效性。算例分析结果表明:该方法能够对时间间隔不同、不等维的序列进行分析,且不需要人为设定聚类算法的参数,户变关系识别准确率高。Factors like low troubleshooting efficiency and untimely data updates make meter-to-transformer wiring relationships in low-voltage distribution networks deviate from the actual situation.To address this issue,a meter-to-transformer relationship identification method based on the combination of angle piecewise linear representation(APLR)and improved clustering by fast search find of density peaks(ICFSFDP)is proposed.Initially,inflection points in the voltage curve are extracted by analyzing the angle variations between neighboring segments,and the curve undergoes adaptive dimensionality reduction and reconstruction using APLR.Then,the ICFSFDP method is deployed to cluster the data sets after dimensionality reduction,and the optimal number of clusters is determined by identifying the minimum area enclosed by the fitted function and the coordinate axis within the decision graph.This allows the identification of central clustered and non-clustered consumers.Finally,the dynamic time warping(DTW)distance is utilized to measure the distance similarity between the central clustered and non-clustered consumers,obtaining meter-to-transformer relationships.The application of this method on both simulated and real data has validated its effectiveness.Results from the analytical cases indicate that this approach can analyze sequences with varied time intervals and dimensions without the need for manually setting clustering algorithm parameters,delivering a high accuracy in identifying meter-to-transformer relationships.

关 键 词：配电网 户变关系 改进分段线性近似 动态时间弯曲距离 最小面积法 密度峰值聚类 

分 类 号：TM743[电气工程—电力系统及自动化]