摘要
针对离散傅里叶变换(DFT)频率估计精度不高的问题,提出了基于DFT输出过零点的系统频率快速估计方法.该方法首先对信号进行DFT输出特性分析,然后取DFT实部(或虚部),利用四阶牛顿插值法精确计算过零点时间,由两个过零点的时间差计算系统频率.仿真分析与实验验证的结果表明,该方法具有较高的频率估计精度,受谐波干扰小,且只需1.78个周波即可实现频率估计,适用于电力系统对实时性要求较高的实际工况.
Aiming at the low accuracy of discrete fourier transform(DFT)frequency estimation,a new fast detection method was proposed based on DFT output zero-crossing.The DFT output characteristics of the signal were analyzed firstly.The zero-crossing time of the DFT real part(or imaginary)was calculated accurately by using the fourth order Newton interpolation method.The system frequency could be obtained by the time difference between two zero-crossing points.The results of simulation analysis and experimental verification showed that the method had higher frequency estimation accuracy and less harmonic disturbance,and only needed 1.78 cycles to achieve frequency estimation,which was suitable for power system with real time requirement.
作者
吴振军
马路遥
邱洪波
WU Zhenjun;MA Luyao;QIU Hongbo(College of Electrical and Information Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,China)
出处
《轻工学报》
CAS
2018年第2期63-69,共7页
Journal of Light Industry
基金
国家自然科学基金项目(51507156)
关键词
频率估计
离散傅里叶变换
过零点法
牛顿插值法
frequency estimation
discrete Fourier transform
zero-crossing method
Newton interpolation algorithm
