基于改进的SVD和Prony的谐波检测算法

Harmonic detection algorithm based on improved SVD and Prony

针对传统的Prony算法在谐波及间谐波检测过程中对噪声敏感,导致辨识精度不高的问题,提出了一种基于奇异值分解(Singular Value Decomposition,SVD)和Prony的改进算法。基于SVD理论,提出了一种奇异点辅助算法,自适应地选取奇异值分解的有效阶次,从而精确地滤除噪声信号。基于已确定的有效阶次,利用改进的Prony算法对滤除噪声之后的信号进行参数辨识,可以准确地估计出各个谐波及间谐波分量的参数。通过MATLAB仿真分析,表明算法能够准确地提取出电力信号的参数信息,具有一定的应用价值。

An improved algorithm based on singular value decomposition(SVD)and Prony is proposed to solve the problem that the traditional Prony algorithm is not very accurate because it is sensitive to noise during harmonic and inter-harmonic detection.Based on SVD theory,a singular point auxiliary algorithm is proposed to select the effective order of singular value decomposition adaptively,so as to accurately filter out the noise signal.According to the exact effective order,the improved Prony algorithm is used to identify the parameters of the signal after noise elimination,and the parameters of each harmonic and its components can be estimated accurately.Through MATLAB simulation analysis,it is shown that the algorithm can accurately extract the parameter information of power signal and has certain application value.