基于大数据的台区行业聚合分类方法及分类特征分析

在研究台区近中期负荷预测方法的过程中，遇到了如何利用大数据识别台区进行行业分类的问题。经过研究，将这个问题分为台区行业分类方法和行业负荷特征两方面。台区行业分类确定了以用电类别作为一级分类，以及运用数据挖掘中的k-means算法对台区典型日年（最大）负荷曲线进行聚类的二级分类共同组成的分类方法；行业负荷特征研究在台区行业分类的基础上，分析行业负荷特征，包括典型日负荷特征和年负荷特征。并以此方法在深圳大数据平台对深圳市台区进行行业分类和分类特征分析。行业分类中将公专变台区一级分类后，都居民生活台区进行聚类分析，分别形成以居民负荷和学校负荷为主的两类。行业负荷特征分析中以学校台区为例，以学生是否住宿为分别，可以区分出走读类学校和住宿类学校。结果表明，此方法效果良好。

Analysis of the Substation Area Industry Clustering Methods and Classification Characteristics Based on the Big Data

In the process of studying the near-medium-term load prediction method of the substation area, we are faced with the problem of how to use big data identification station area for industry classification. After the research, this problem is divided into the substation area industry classification method and industry load characteristics. The industry classification of the substation area determines the classification method of the second-level classification of the typical daily year (maximum) load curve in the substation area by using the k-means algorithm in data mining, and the industry load characteristic study analyzes the industry load characteristics on the basis of the industry classification of the substation area. Includes typical daily load characteristics and annual load characteristics. And this method in Shenzhen big data platform for the Shenzhen substation area industry classification and classification characteristics analysis. After the classification of the public-specific change substation area district in the industry classification, the residents'' living area is analyzed, and two categories are formed, respectively, based on the resident load and the school load. Industry load characteristics analysis to take the school substation area district as an example, to whether the students accommodation is separate, you can distinguish between out-of-school schools and residential schools. The results show that this method works well.