基于轨迹灵敏度和FASTICA的同调机群识别方法  

Method of Identifying Coherent Generator Groups Based on Trajectory Sensitivity and FASTICA

作　　者：王永贵[1] 卫志农[1] 孙国强[1] 何桦[2] 

机构地区：[1]河海大学可再生能源发电技术教育部工程研究中心,南京210098 [2]南瑞继保电气有限公司,南京211100

出　　处：《电力系统及其自动化学报》2014年第3期14-19,共6页Proceedings of the CSU-EPSA

基　　金：国家自然科学基金资助项目(51277052;50977021;51107032)

摘　　要：追求准确、快速地识别同调机群对于研究电力系统稳定性具有重要意义。提出了一种基于发电机近似摇摆曲线轨迹灵敏度和快速独立分量分析FASTICA(fast independent component analysis)的同调机群识别方法。首先求得发电机的近似摇摆曲线,然后对其轨迹求灵敏度,最后运用盲信号源分解问题中的FASTICA技术对发电机摇摆曲线灵敏度数据进行同调机群识别,有效提高分群的准确性。通过对IEEE 39节点、IEEE145节点和某省级电网的仿真计算验证了该方法的有效性和可行性。Pursuiting of accurately and quickly identifying coherent generator groups is of great significance in studying the stability of the power system. A method based on the trajectory sensitivity of the approximate swing curves of the disturbed generators and FASTICA is proposed for recognizing coherent generator groups in power systems. Firstly, the approximate swing curves of the disturbed generators are obtained, and then the trajectory sensitivity is calculated, fi- nally, the FASTICA for blind source decomposition problem is used for coherency recognition with the trajectory sensi- tivity data of the swing curves. The coherent generator groups are obtained in accordance with the trend of the dynamic changes of the generator rotor angle. The accuracy of clustering is improved effectively. The validity of the method is verified by the simulation of the IEEE 39-bus, IEEE 145-bus system and a provincial grid.

关 键 词：电力系统 同调机群 轨迹灵敏度 快速独立分量分析 

分 类 号：TM743[电气工程—电力系统及自动化]