基于矩形卷积窗算法的电子式电流互感器在线校验系统

发布时间：2018-05-27 21:06

  本文选题：互感器 + 虚拟仪器　； 参考：《电网与清洁能源》2015年11期

【摘要】：鉴于传统的FFT(快速傅里叶)算法仅适用于整周期采样后的离散信号,电流互感器校验系统硬件容易混入噪声等问题,基于矩形卷积窗算法设计了一种可对电子式电流互感器进行不停电校验的在线校验系统。介绍了校验系统的整体结构,采用FFT加四重矩形卷积窗的高精度插值算法,并设计了一种复合滤波法程序以消除随机误差。其中,设计的虚拟校验系统可以有效降低硬件成本,避免硬件中可能混入的噪声,可以针对现场运行的电子式电流互感器进行在线校验,更精确地反映一次侧电流电压值,提高了电力系统的可靠性。实验结果表明:相对于传统的FFT算法的校验系统,所设计基于矩形卷积窗算法的虚拟校验系统有更高精度。
[Abstract]:Since the traditional FFT (Fast Fourier transform) algorithm is only suitable for discrete signals after the whole period sampling, it is easy to mix noise into the hardware of the current transformer calibration system. Based on the algorithm of rectangular convolutional window, an online calibration system is designed to check the failure of the electronic current transformer. This paper introduces the whole structure of the checking system, adopts the high precision interpolation algorithm of FFT and quadrilateral convolution window, and designs a compound filter program to eliminate the random error. The designed virtual calibration system can effectively reduce the cost of hardware and avoid the possible noise in the hardware. It can be used to calibrate the electronic current transformer running in the field and reflect the primary side current and voltage more accurately. The reliability of power system is improved. The experimental results show that the virtual verification system based on the rectangular convolution window algorithm is more accurate than the traditional FFT algorithm.
【作者单位】： 湖北宜昌供电公司;三峡大学电气与新能源学院;
【基金】：国家自然科学基金项目(51507091)~~
【分类号】：TM452
，

本文编号：1943746