一种快速的软件频率计算方法  

A fast software frequency calculation method

作　　者：李肖博 姚浩 于杨 习伟 蔡田田 LI Xiaobo;YAO Hao;YU Yang;XI Wei;CAI Tiantian(China Southern Power Grid Digital Grid Research Institute Co.,Ltd.,Guangzhou 510640,China)

机构地区：[1]南方电网数字电网研究院有限公司,广东广州510640

出　　处：《电力系统保护与控制》2022年第1期181-187,共7页Power System Protection and Control

基　　金：中国南方电网有限公司科技项目资助(ZBKJXM 20180500);国家重点研发计划资助(2017YFB0904900,2018YFB0904902)。

摘　　要：频率测量是电力系统测控装置和保护装置的基本功能。频率测量的快速性和准确性将深度影响电力系统测控和保护装置的性能,从而影响到电力系统的稳定性。为提高频率测量的快速性,提出了一种全新的频率测量算法。该算法可以在5/4周波内测量出被测对象的频率,具有极强的实时性。首先对计算傅里叶级数实部的核函数进行移相。然后分别采用移相前后的核函数对原始信号进行傅里叶变换,求取实部信号。接着利用两个实部信号过零点的时间差计算出信号频率。最后使用Matlab仿真和保护装置实测对该算法进行了验证。验证结果显示:所提测频算法对频率的测量满足精度要求;相对于现有算法,对渐变频率的跟踪速度有较大的提高。Frequency measurement is the basic function of power system measurement and of a control device and protection device. The rapidity and accuracy of frequency measurement will seriously affect the performance of power system measurement and the protection device, thus affecting the stability of the power system. In order to improve the rapidity of frequency measurement, a new frequency measurement algorithm is proposed. The algorithm can measure the frequency of the measured object in 5/4 cycle, and has strong real-time performance. First, the kernel function for calculating the real part of a Fourier series is phase shifted. Then the kernel function before and after phase shift is used to Fourier transform the original signal to obtain the real part of the signal. After that, the signal frequency is calculated by using the time difference between the zero crossing of two real part signals. Finally, the algorithm is verified by Matlab simulation and actual measurement of a protection device. The tests show that the proposed algorithm meets the accuracy requirements;the tracking speed of gradually changed frequency is greatly improved compared to that of existing algorithms.

关 键 词：频率测量 傅里叶级数 核函数 移相 MATLAB 

分 类 号：TM935.1[电气工程—电力电子与电力传动]