基于聚类融合技术的电力用户负荷模式提取方法

针对电力大数据背景下智能电力用户负荷模式提取的可靠性不高且传统单一聚类算法聚类结果不稳定的问题,提出一种基于主成分分析与聚类融合相结合的电力用户负荷模式提取方法。首先,对负荷数据进行预处理,通过主成分分析法减少特征间分类信息冗余实现高维特征的降维。然后,用四种聚类方法分别对降维后的数据集进行聚类分析,得到具有差异性的聚类成员。最后,利用共识矩阵对所得聚类成员进行聚类融合,得到优于单一聚类算法的最终聚类结果。通过电网实际用电数据验证了所提负荷模式提取方法能够提高聚类准确率并降低计算复杂性,并用有效性指标Silhouette对最终聚类结果进行评价。

Electricity customer load pattern extraction based on clustering ensemble approaches

To solve the problem that the clustering results of traditional single clustering algorithm is unsteady and the reliability of load pattern extraction is low, a new method of pattern classification based on the technology of principal component analysis and clustering ensemble approaches is presented. First, principal component analysis is used in the process of data preprocessing to reduce information redundancy of high-dimension feature vectors. Then, different clustering members are obtained by four cluster analyses of feature vectors. Finally, the clustering members are combined with the Co-association Matrix and the clustering result is better than single clustering algorithm. The clustering results of principal component analysis and ensemble clustering algorithm achieved by electricity consumption data show that it can effectively improve the accuracy of clustering and reduce computation complexity. The evaluation indicator of Silhouette is used to assess the clusters of load profiles.