基于改进FFT的合并单元计量性能校验算法研究

采用常规FFT算法作为合并单元计量性能校验仪的算法时,由于频谱泄漏和栅栏效应,比差和相差不满足0.05级的准确度要求。针对这一问题,在常规FFT的基础上,提出了加Hanning窗的单谱线插值FFT算法和加Blackman窗的双谱线插值FFT算法这两种改进FFT算法,在LabVIEW平台上对这两种算法和常规FFT算法进行仿真比较,仿真结果表明,加Blackman窗的双谱线插值FFT算法的计算准确度最高,更适合用于合并单元计量性能校验。最后在国家高压计量站对合并单元计量性能校验仪进行了校准试验,测试数据表明校验仪满足0.05级准确度要求,从而验证了加Blackman窗的双谱线插值FFT算法的有效性和可行性。

Improved FFT Algorithm Based Merging Unit Metering Performance Calibration Algorithm

Fast Fourier transform (FFT), usually suffering from the problem of frequency spectrum leakage and picket fence effect, would not be suitable as the calibration algorithm of merging unit metering characteristics calibration device, so the ratio error and phase displacement of the calibration device utilizing common FFT algorithm cannot satisfy the requirement of 0.05 accuracy class specification.This paper, based on common FFT, proposes two improved FFT algorithms, i.e., single line interpolation FFT algorithm with Hanning window and two-line interpolation FFT algorithm with Blackman window, these two algorithms are compared with the common FFT algorithm on LabVIEW platform, the simulation results manifest that the two-line interpolation FFT algorithm with Blackman window has the highest accuracy for harmonic analysis, which is more suitable for merging unit metering performance calibration.Finally, a prototype calibration device using this algorithm is calibrated at National High Voltage Metering Station, and the results show that the device can meet the specification of 0.05 accuracy class, confirming the effectiveness and feasibility of the two-line interpolation FFT algorithm with Blackman window.