Research On Method And Application Of Improved Quasi-synchronous Sampling Algorithm For Power Frequency Measurement

Frequency, as we all know, is an key parameter of power quality, which is of great importance for power system monitoring, control, and protection. With synchronous sampling, accurate frequency measurement can be easily achieved by using the Fast Fourier Transform(FFT). However, the fluctuations of the fundamental frequency and harmonics in power system makes synchronous sampling unattainable.In the case of asynchronous sampling, the performances of FFT-based frequency measurement have the inherent limitations such as spectral leakage and picket fence effects which lead to incorrect results of frequency measurement. Thus, how to overcome the impact of non-synchronous sampling to frequency deviation on frequency measurement has become a hot topic for years. This paper presents an improved quasi-synchronous sampling algorithm(IQSSA) in order to solve the aforementioned problems which successfully increases the frequency measurement accuracy in case of non-synchronous sampling.Firstly, various frequency measurement algorithms and the basic principles of quasi-synchronous sampling algorithm(QSSA) are introduced. The QSSA can obtain the interested spectrum of each sampled sequence by a simple iteration procedure based on the time domain integration. However, the major shortages of the QSSA are the high computation and low accuracy.Secondly, the improved quasi-synchronous sampling algorithm based on the numerical integration is proposed where the composite trapezoidal integration and composite Simpson integration are adopted. The workflows of IQSSA-based frequency measurement by using the composite trapezoidal integration and composite Simpson integration are proposed. Comparisons of computational burden between the windowed interpolation FFT(WIFFT) and IQSSA are presented. It shows that the computation of frequency estimation obtained by the IQSSA is lower than that of WIFFT when the same numbers of samples are used. Moreover, IQSSAs based on the composite trapezoidal integration and composite Simpson integration have the same time complexity, which is because the time complexity is commonly estimated by counting the number of elementary operations performed by the algorithm.Thirdly, simulations are performed to illustrate the accuracy of frequency measurement of the proposed IQSSA with comparisons to the WIFFT based on the four-item and third-order of Nuttall window(NW 4-III). The influence of fundamental frequency fluctuations, harmonic interference, and white noise on frequency measurement results are evaluated. The simulation results show that the accuracy of frequency measurement provided by the IQSSA are higher than that of the WIFFT based on the NW 4-III. Especially, the IQSSA has much better performance than that of WIFFT based on the NW 4-III with the influence of white noise. Moreover, the variances of frequency measurement provided by the IQSSA based on the composite trapezoidal integration are less than that of the IQSSA based on the composite Simpson integration.Finally, in order to verify the proposed IQSSA, the laboratory experiments are done. The calibration scheme for the power system frequency measurement is proposed according to the national standards, i.e., GB/T 15945-2008, GB/T 19862-2005, and DL/T 1028-2006. The measurement uncertainty of the frequency measurement results is analyzed. The experiment results show that the proposed method can achieve the same level of accuracy as the method recommended in the national standards. Thus both simulation and experimental results have verified the effectiveness and accuracy of the IQSSA in the power system frequency measurement.