基于Matrix Pencil的低频振荡辨识及PSS优化配置研究  

Research on Low-frequency Oscillation Identification Based on the Matrix Pencil Algorithm and PSS Optimal Allocation

作　　者：张军财[1] 金涛[1] 

机构地区：[1]福州大学电气工程与自动化学院,福州350108

出　　处：《电气技术》2013年第1期18-22,共5页Electrical Engineering

基　　金：欧盟FP7国际科技合作基金(909880);国家自然科学基金(50907011);福建省自然科学基金(2010J05109);教育部留学归国人员基金;福建省杰出青年科学基金

摘　　要：在互联系统中,低频振荡问题越来越成为影响电力系统稳定的重要因素。广域测量系统(WAMS)的发展和应用使得低频振荡的在线辨识和PSS配置及参数优化成为可能。本文探讨了在已知干扰情况下,利用测量数据构造矩阵,基于Matrix Pencil算法来进行电力系统低频振荡模式分析,建立多输入输出系统的低阶近似的辨识传递函数,并在此基础上利用奇异值分解(SVD)的方法来分析PSS优化配置问题。通过经4机2区域系统的仿真实验研究表明,基于Matrix Pencil辨识算法能建立准确的低阶近似传递函数,奇异值分解的方法能有效分析出PSS最佳配置地点。In the interconnected power system,the impact of low-frequency oscillation is increasingly becoming an important factor for the power system stability problem.The development and application of Wide Area Measurement System（WAMS） make on-line identification of low-frequency oscillations mode and PSS configuration and parameter optimization possible.Under the condition of known input,this thesis discusses to structure matrixes basing on the measurement data,and use Matrix Pencil algorithm to analyze the oscillation mode of power system low frequency,and then establish identification of transfer function of a multi-input and output system;On this basis,the paper further discusses to use the singular value decomposition（SVD） method to analyze the PSS optimal allocation problem.Through the simulation experiment research of the example of a two regional,four machines system,it shows that the identification algorithm based on the Matrix Pencil can establish an accurate low dimensional approximation of the transfer function,and singular value decomposition（SVD） method can effectively analyze the PSS optimal allocation.

关 键 词：广域测量 低频振荡 Matrix PENCIL 辨识传递函数 奇异值分解 PSS配置 

分 类 号：TM712[电气工程—电力系统及自动化]