一种基于自回归模型的间谐波谱估计的改进算法  

AN IMPROVED ALGORITHM OF INTERHARMONIC SPECTRAL ESTIMATION BASED ON AR MODEL

作　　者：马秉伟[1] 刘会金[1] 周莉[1] 崔福鑫[1] 

机构地区：[1]武汉大学电气工程学院,湖北省武汉市430072

出　　处：《中国电机工程学报》2005年第15期79-83,共5页Proceedings of the CSEE

摘　　要：间谐波的频率是基频的非整数倍,频谱随时间而变化,具有随机性,因此间谐波是随机信号。该文首先介绍平稳随机信号的AR模型和Burg算法的基本原理,然后分析Burg算法在处理正弦信号时出现谱峰偏移的原因。在此基础上该文提出了一种基于AR模型的间谐波谱估计改进算法。该算法通过直接求解在预测误差功率最小意义下的较低阶AR模型系数,再递推计算高阶系数,减小了谱估计的谱峰偏移。仿真结果表明,该算法明显改善了谱估计的性能,而且只要用比较短的数据即可得出较好的间谐波谱估计。The frequency of interharmonics is not an integer of the fundamental, and it＇s spectral is changeful. So, interharmonics is random signal. This paper introduces the AR model of stationary random signal and the principle of Burg algorithm, analyses the peak shifting in the case of processing sinusoidal signal, and proposes an improved algorithm of interharmonic spectral estimation based on AR model. The algorithm improves peak shifting of parameter spectral estimation through solving the coefficients of the low order AR model with the meaning of the minimum prediction error power, then gets the higher order coefficients. The simulating result shows that the algorithm improves spectral estimation performance obviously, and can get better interharmonic spectral estimation with less datas.

关 键 词：电能质量 间谐波 自回归模型 Burg算法 

分 类 号：TM711[电气工程—电力系统及自动化] TM76