基于DFT的周期信号谐波特性测试与仿真研究

Analysis of Periodic Signal Harmonic Characteristics Based on DFT

周期信号谐波分析是获取构成信号中各种频谱分量的幅度和相位的处理过程,被广泛的应用于电能质量分析、故障诊断、语音分析等领域。DFT谐波分析方法在实际应用中,由于DFT固有的特点,易于造成频谱泄漏而影响分析结果。从谐波分析的原理出发,研究了基于DFT实现周期信号谐波分析的机理和频谱泄漏产生的原因,指出了提高谐波特性分析准确性的方向。理论分析和仿真结果证明,正确的选择采样频率、选择数据处理的相关参数是关键,仅当进行FFT处理的有限长序列进行周期延拓后,与原采样序列相同时,将不会产生频谱泄漏,谐波分析结果正确。

Harmonic analysis of periodic signals is to obtain the amplitude, frequency and phase of sine components formed the signals, which has been widely used in the field of power quality analysis, fault diagnosis, and speech analysis. Due to the characteristics of digital signal processing, it is easy to cause spectral leakage and affect analysis results. In this paper, theoretical analysis and experimental research are implemented aiming at the procedure of harmonic analysis based on DFT under single-cycle and multi-cycle uniform sampling modes. The results show that that proper parameters of sampling and data processing are critical, it would not produce spectral leakage and have correct harmonic analysis results, only when cycle extension of the finite sequence performed by FFT is the same as the original sample sequence.