基于特征聚类分析的大规模发电数据异常辨识方法

An Abnormal Data Identification Method of Large-scale Generation Data Based on Cluster Analysis

摘要: 目前电厂发电量数据集抄已在各电网公司得到了广泛应用。针对电量集抄中首要解决的异常数据辨识问题，深入分析了不同类型电源的自身特性，基于特征聚类分析技术，提出了不同类型电源日发电量异常的辨识条件，由此构建了基于特征聚类分析的大规模发电数据异常辨识方法，最后基于某省网的实际数据，验证了所提出方法的有效性。

Abstract: Centralized collection technology of generation data is widely used in many power grid companies at present. To solve the abnormal data identification problem, this paper analyzed the different type of power generation. Then an abnormal data identification method of large-scale generation data based on cluster analysis was proposed. At last, the effectiveness of the proposed method was verified based on actual data from a province grid company.