基于频谱分析的负荷形态分类研究

摘要

In order to do better with the economic analysis, operation and planning of the power system, we need to seek for an efficient and reliable method of load morphological classification. We took spectrum analysis on daily average load data of power us-ers, and classified all users into several categories through dynam-ic clustering method. The result of example showed that the inte-grated cluster validity of frequency-domain method was higher than that of time-domain method. The frequency-domain method improves traditional time-domain method and lays a solid founda-tion for utilities to carry on accurate price setting, load forecasting and orderly power utilization.%为了更好地对电力系统进行经济分析、运行与规划,需要寻求高效可靠的负荷形态分类方法.对电力用户日平均负荷数据进行频谱分析,采用动态聚类法将不同的电力用户负荷形态进行聚类并将此方法与传统时域上的负荷形态分类方法比较,算例结果表明,基于频域分析的负荷形态分类方法综合聚类有效性更高,其改善了负荷特性时域分析的不足,为电力公司准确进行电价定制、负荷预测及制定有序用电决策方案提供参考.