摘要
Burg算法适合于电力系统间谐波谱估计,但存在谱峰偏移和谱线分裂的缺点,影响其谱估计效果。本文在误差分析的基础上提出了基于最优窗Burg算法的间谐波谱估计新方法。首先通过在平均频率误差方差最小的意义下来获得最优窗,再将加窗后的预测误差平均功率最小化来求取反射系数,最终求得信号的功率谱。仿真结果表明,与传统的Burg算法和汉明窗Burg算法相比,最优窗Burg算法具有更好的谱估计性能。并且,由于采用Levinson递推,其计算复杂度明显低于特征值法。该方法可应用于电力系统间谐波谱估计。
The Burg algorithm is suitable for spectral estimation of inter-harmonic in power system,but its effect of spectral estimation is affected by peak bias and spectral line splitting.By analyzing the error,a new method of inter-harmonic spectral estimation based on the optimal window Burg algorithm is presented in this paper.The optimal window can be obtained by minimizing the variance of average frequency error.Then reflection coefficient is obtained by minimizing the average power of windowed prediction error.Finally,the power spectrum is obtained.Experimental results show this method has high frequency resolution compared with Burg algorithm and hamming window Burg algorithm.And by Levinson iteration,this method has lower computational complexity than eigenvalue method.This method can be used in the inter-harmonic spectral estimation in power system.
出处
《电工技术学报》
EI
CSCD
北大核心
2011年第1期177-182,共6页
Transactions of China Electrotechnical Society
关键词
电能质量
最优窗Burg算法
间谐波
电力系统
Power quality
the optimal window Burg algorithm
inter-harmonics
power system
