基于六项余弦窗四谱线插值FFT的高精度谐波检测算法

目前，在实际电网环境中谐波检测算法的精确度不够高。为此，分析了一种具有旁瓣峰值低且下降速度快的六项余弦窗并将其应用于FFT算法中，提出了基于六项余弦窗四谱线插值FFT的谐波检测算法，运用多项式拟合polyfit函数推导出了简单实用的四谱线校正公式，简化了运算过程。实验结果表明:该算法在21次复杂谐波环境中，与四项Nuttall窗和四项Rife-Vincent窗FFT插值相比有更高的精确度，且在基于相同窗的情况下，四谱线插值的精确度要高于三谱线插值计算。最后，在实验中考虑到实际电网中可能遇到的噪声干扰情况，进行了仿真，仿真结果验证了该算法具有较高的检测精确度。

An accurate approach for harmonic detection based on 6-term cosine window and quadruple-spectrum-line interpolation FFT

The accuracy of harmonic detection in the real power grid is not high enough at present. So a 6-term cosine window with maximum sidelobe decay which is applied into the FFT algorithm is analyzed, and an improved FFT method for harmonic detection based on 6-term cosine window and quadruple-spectrum-line interpolation is proposed. The applicable rectification formula of the quadruple-spectrum-line interpolation is obtained by using polynomial curve fit functions so that calculating burden is reduced. The simulation results show that in the algorithm the analysis precision is better than 4-term Nuttall window and 4-term Rife-Vincent window interpolated FFT in the 21 orders harmonics. Simultaneously, the precision of quadruple-spectrum-line interpolation is also better than that of triple-spectrum-line interpolation correction algorithm when the same window is used. At last the results of experiment verify that detection accuracy will be kept high when facing the noise interferences that may be encountered in real grid. This work is supported by The Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (No. KJ2015QT007).