基于模糊聚类的区域变电站节电划分方法

Electricity-Saving Classification Method of Area Substations Based on Fuzzy Clustering

区域变电站具有数目众多、类型多样、线损构成及节电特性复杂的特点。为提升区域变电站节电研究效率,实现区域变电站节电特性的科学划分,文中提出了一种基于模糊c-均值聚类(FCM)的变电站节电划分的方法.首先,设计一套变电站节电划分综合指标体系,包括涵盖变电站主变容量、主变台数等基本参数的基本属性指标子体系和涵盖输送电量、站用电量等电气参数的电气属性指标子体系;其次,依据各指标对变电站的损耗及其节电特性的影响程度,应用层次分析法(AHP)确定各指标权重,建立AHP-FCM的变电站节电划分模型;然后,分别应用所设计的节电划分综合指标体系、基本属性指标子体系和电气属性指标子体系从3个不同的角度运用AHP-FCM实现区域变电站的划分,通过对3种节电划分结果及其聚类中心矩阵分析研究,获得对各类变电站损耗及其综合节电特性的深入认识,为变电站后续的研究提供参考依据和重要指导;最后,以某区域90个变电站的实际数据进行实例分析,验证了所提方法的实用性和有效性.

Area substations are numerous and diverse, and they have complex line loss and electricity-saving fea- tures. In order to effectively investigate the electricity-saving of area substations and scientifically classify area sub- stations, an electricity-saving classification method of area substations is proposed based on the fuzzy c-means clus- tering. In the method, first, a comprehensive index system is designed for the electricity-saving classification, which includes the basic property index sub system and the electrical property index sub system. The basic property index sub system covers the basic parameters of substations, such as the capacity and number of main transformers. The electrical property index sub system covers the electrical parameters of substations, such as the transport capa- city and the electricity consumption. Next, according to the influence of each index on the loss and electricity-sav- ing features of the substations, the index weights are determined by means of the analytic hierarchy process （AHP） , and an electricity-saving classification model of the substations is constructed based on AHP-FCM. Then, numerous area substations are classified into several categories respectively by utilizing the designed comprehensive index system, the basic property index sub system and the electrical property index sub system. Moreover, the in- depth understanding of the loss and comprehensive electricity-saving features of the substations are obtained by ana- lyzing and comparing three kinds of electricity-saving classification results and the corresponding clustering results, which can provide guidance for the follow-up studies. Finally, the feasibility and effectiveness of the proposed method are verified by analyzing the actual data of 90 transformer substations.