基于关联特征筛选的双层聚类区域配变电压越限特征分析及预测  

Analysis and Prediction of Over-Limit Characteristics for Regional Distribution Transformer Voltage Using Bilayer Clustering by Correlation Feature Screening

作　　者：郭少东 赵晓莉 孙改平 杨秀 杨帆 刘俊 GUO Shaodong;ZHAO Xiaoli;SUN Gaiping;YANG Xiu;YANG Fan;LIU Jun(School of Electrical Engineering,Shanghai University of Electric Power,Shanghai 200090,China;State Grid Shanghai Electric Power Research Institute,Shanghai 200437,China)

机构地区：[1]上海电力大学电气工程学院,上海200090 [2]国家电网上海电力科学研究院,上海200437

出　　处：《南方电网技术》2025年第2期19-27,共9页Southern Power System Technology

基　　金：国家自然科学基金资助项目(52207121);上海电力人工智能工程技术研究中心项目(19DZ2252800)。

摘　　要：针对区域配网变压器(简称“配变”)数量多,大量新型负荷、分布式光伏等接入,配变电压随机性波动增强,台区用户电压质量面临挑战。为更好地对区域配变电压进行越限特征分析及预测,提出了基于关联特征筛选的双层聚类区域配变电压预测方法。首先,将区域配变的越限天数作为第一层聚类特征,获得电压性质正常以及越上限的配变。其次,针对电压越限配变提出结合Pearson相关系数和欧氏距离(Euclidean distance)的最优度量矩阵,提取原有数据的内含信息,作为K均值聚类(K-means)的输入,实现对区域配变双层聚类。在此基础上,选取该集群中代表配变表征该类配变,利用卷积双向长短期记忆网络-注意力机制(convolutional neural network-bidirectional long and short-term memory-attention,CNN-BiLSTM-Attention)模型对配变电压进行预测,该模型能够提取输入数据的双向信息特征,并对重要特征加权,从多时间尺度上获得双向特征信息用于预测。最后,在上海市某区域配变验证了该方法的有效性。Aiming at the large number of regional distribution transformer,a large number of new loads,distributed photovoltaics,etc.,and the enhancement of the random voltage fluctuation of distribution transformer.The voltage quality of the substation users is facing challenge.In order to better analyze and predict the over-limit characteristics of regional distribution transformer voltage,a bilayer clustering regional distribution transformer voltage prediction method based on correlation feature screening is proposed.Firstly,the number of overrun days of regional distribution transformers are taken as the first layer clustering feature,and the distribu-tion transformers with normal and over-limit voltage properties are obtained.Secondly,for the over-limit voltage distribution transformers,an optimal metric matrix combining Pearson′scorrelation coefficient and Euclidean distance is proposed to extract the contained information of the original data as the input of K-means to realize the bilayer clustering of regional distribution transformer.On this basis,the representative distribution transformers in the cluster are selected to characterize the distribution transformers of this category,and the convolutional neural network-bidirectional long and short-term memory-attention(CNN-BiLSTM-Attention)model is used to predict the distribution transformer voltage,which can extract the bidirectional information features of the input data,weight the important features,and obtain the bidirectional feature information from multiple time scales for prediction.Finally,the effectiveness of the proposed method is verified in a certain area of Shanghai.

关 键 词：区域配变 最优度量矩阵 双层聚类 降维 电压预测 

分 类 号：TM711[电气工程—电力系统及自动化] TM714