基于卷积高斯窗的电能计量研究

摘要

In this paper,electrical energy measurement analysis is realized by combining convolution Gauss window method and Fourier transform.Taking advantage of the best time-frequency resolution of Gauss window function,the proposed method can accurately identify different harmonic components in electrical signals and complete the estimation of electrical power.The experimental simulation containing an electrical signal of 5 harmonic components compares frequency,amplitude and phase estimation errors of different orders of harmonics by Gauss Window function and convolution Gauss Window function to verify that the second convolution Gauss window function has the best performance,where the relative estimation errors of electrical signal parameters are all below 0.1‰.%结合卷积高斯窗和傅里叶变换实现电能计量分析,利用高斯窗函数良好的时频分辨能力,可以准确辨识电能信号中不同的谐波分量,完成电能精确估算.实验仿真电能信号含有5阶谐波分量,对比了高斯窗函数,卷积高斯窗函数对仿真电能信号不同阶次谐波分量的频率、幅度和相位的数值估计误差,验证了卷积两次的高斯窗函数的性能最优,其对于电能信号谐波参数的估计相对误差均在0.1‰以下.