基于自适应神经网络系统聚类的特殊负荷分类研究

Research on Special Load Classification Based on Self-Adaptive Neural Network Hierarchical Cluster

根据特殊负荷调查统计信息,甄别不同聚类分析的优缺点,提出了一种基于自适应神经网络系统聚类的特殊负荷分类方法,分别从负荷特性、供电可靠性与电能质量3个维度建立聚类指标,并以78个特殊负荷为样本进行聚类分析。结果表明,该方法快速有效,同类特殊负荷相似度高,各类特殊负荷之间差异明显。该分类方法可推广至典型特殊行业负荷分类,为配电网对特殊负荷采用相应的供电措施提供理论依据,便于供电部门制定特殊负荷供电服务。

Based on the information of the special load surveyed,by identifying advantages and disadvantages of different cluster analyses,a special load classification method based on self-adaptive neural network hierarchical cluster is proposed in this paper.Clustering indices are established from three dimensions of load characteristics,power supply reliability and power quality,and 78 special loads are taken as samples for the cluster analysis.The results show that the proposed method is fast and effective,the similarity of similar special loads is high,and differences between different kinds of special loads are obvious.This classification method can be extended to the load classification of typical special industries,providing theoretical basis for the distribution network to adopt corresponding power supply measures for special loads,and facilitates the power supply company to formulate power supply services for special loads.