Data-Driven Mode Identification Method for Broad-Band Oscillation of Interconnected Power System
Data-Driven Mode Identification Method for Broad-Band Oscillation of Interconnected Power System
复制标题
DOI：
10.1109/jsen.2022.3187109
复制
发表时间：
2022
影响因子：
4.3
通讯作者：
Yong Li
中科院分区：
文献类型：
--
作者：
Fang Liu;Sisi Lin;Junjie Ma;Yong Li
The paper presents research on mode identification of broad-band oscillation in interconnected power system. A data-driven mode identification (DDMI) method for broad-band oscillation signals is proposed creatively in this paper. Firstly, piecewise aggregation approximation algorithm is improved to achieve effective dimension reduction of oscillation data. Combined with ${k}$ -Shape clustering algorithm, oscillation database is established with historical data, real-time data and simulation data. Then, oscillation mode identification models corresponding to different data categories can be obtained based on random forest algorithm, which can realize fast and automatic matching between broad-band oscillation data and oscillation mode parameters. Finally, the identification results of two simulation oscillation cases and an actual oscillation case show that proposed method can accurately identify the oscillation mode parameters from broad-band oscillation signals and has higher accuracy compared with other methods.
