模糊ISODATA算法在城市负荷分类中的应用

鉴于目前电力大数据总量大、数据复杂度高、价值密度低等,基于模糊ISODATA聚类算法研究了商场、宾馆酒店、办公等典型城市负荷用电特性,建立了城市负荷的建筑类型、总能耗、照明与插座用能、空调用能、动力用能、特殊和其他用能的数据维度模型用以辅助分类决策,并在不同维度、数量级的样本数据下与传统聚类算法的性能进行比较。同时与传统的基于日负荷曲线的分类方法准确率进行对比。结果表明,基于模糊ISODATA算法的城市用电行为分析模型聚类效果好、精度高,适用于当前高维度、大规模的电力大数据的分类,并在一定程度上能够克服当前电力大数据负荷分类中聚类数目设置难、易陷入局部最优等问题。研究成果可用于指导工程实践。

Application of Fuzzy ISODATA Algorithm in Urban Load Classification

In order to deal with the problems of intelligent data mining, such as large data volume, high complexity, and low mining efficiency, this paper does research on the electricity consumption of urban power grid, and classifies typical urban load such as shopping malls, hotels, office. A data dimension model of building types, total energy consumption, lighting and sockets, air-conditioning energy, power-using energy, special and other energy use is established. It is compared with the traditional classification effect based on the daily load curve. The results of fuzzy ISODATA clustering are more reasonable and effective, and it is applicable to the classification of current high-dimensional, multi-complex and related power big data, and can overcome the problems of setting number of clusters and falling into local optimum in current electric load classification of big data. Results of the research can be used to guide engineering practice.