改进准同步采样电力系统频率测量方法及应用研究

发布时间：2018-09-06 16:12

【摘要】：电力系统频率是电能质量的重要指标,其准确测量有助于客观反映电力系统的运行状态,具有重要的理论和实际意义。同步采样情况下,采用快速傅里叶变换(Fast Fourier Transform,FFT)可以获得高准确度的频率测量结果。然而,在电力系统中,信号频率往往存在一定波动,且受到谐波等干扰的影响,使严格的同步采样很难实现。在非同步采样条件下,FFT固有的频谱泄漏和栅栏效应影响了频率测量的准确度。因此,如何克服非同步采样偏差对频率测量结果的影响,一直是国内外学者研究的重点。本文提出并研究改进准同步采样电力系统频率测量方法,提高了非同步采样情况下的频率测量精度,为电力系统频率测量方法的发展与应用提供了新思路。本文首先综述了常用频率测量方法的优缺点,介绍了准同步采样算法的基本原理。准同步采样算法通过积分和迭代运算实现局部信号的频谱计算,且允许信号存在一定的非同步采样偏差,但该方法也存在计算量较大、准确度不高等不足。其次,本文根据准同步采样算法实现过程中需要进行数值积分的特点,在基于复化矩形的准同步采样算法推导过程的基础上,研究并推导了改进准同步采样频率测量算法,包括基于复化梯形的准同步采样频率测量算法和基于复化Simpson的准同步采样频率测量算法,建立了改进准同步采样电力系统频率测量实现流程,从理论上对比分析了常用的加窗插值FFT算法和改进准同步采样频率测量算法的运算复杂度。本文研究结果表明:当采样频率和采样点数相同时,改进准同步采样频率测量算法执行过程中所需加法次数远远少于加窗插值FFT算法所需加法次数。由于基于复化矩形、复化梯形和复化Simpson三者仅在积分加权系数上不同,且准同步采样频率测量算法的运算复杂度仅与采样点数有关,因此本文所研究的两种改进准同步采样频率测量算法的运算复杂度相同。随后,本文通过仿真实验对比分析了改进准同步采样频率测量算法与加窗插值FFT算法的测量准确度。本文选取了被广泛应用的基于4项3阶Nuttall窗的插值FFT算法作对比,仿真分析了基波频率波动、谐波干扰、白噪声影响等不同情况下的频率测量结果。仿真结果表明:与基于4项3阶Nuttall窗的插值FFT算法相比,本文所研究的改进准同步采样频率测量算法具有更高的频率测量准确度,且该算法可有效抑制白噪声对电力系统频率测量的影响。在两种改进准同步采样频率测量算法中,基于复化梯形的准同步采样频率测量算法的性能略优于基于复化Simpson的准同步采样频率测量算法。最后,本文利用嵌入式系统平台实现了本文所提出的改进准同步采样频率测量算法,并根据GB/T 15945-2008、GB/T 19862-2005、DL/T 1028-2006的要求设计了电力系统频率测量校准方案。通过大量校准前后频率测量实验,对比分析了校准前后的电力系统频率测量偏差,并分析了基波频率不变时校准后的电力系统频率测量不确定度。实验结果表明:校准后的频率测量偏差均能达到国标规定的频率偏差在±0.01Hz以内的要求,这也验证了本文所提出的改进准同步采样频率测量算法在实际应用中的的有效性和准确性。
[Abstract]:Power system frequency is an important index of power quality. Its accurate measurement is helpful to reflect the operation state of power system objectively. It has important theoretical and practical significance. Under the condition of asynchronous sampling, the inherent spectrum leakage and fence effect of FFT affect the accuracy of frequency measurement. Therefore, how to overcome the influence of asynchronous sampling deviation on frequency measurement results has been a problem at home and abroad. This paper presents and studies the improvement of frequency measurement method of quasi-synchronous sampling power system, which improves the accuracy of frequency measurement under the condition of non-synchronous sampling. It provides a new idea for the development and application of frequency measurement method of power system. The quasi-synchronous sampling algorithm realizes the spectrum calculation of local signal by integral and iterative operation, and allows the signal to have some asynchronous sampling deviation, but this method also has the shortcomings of large amount of calculation and low accuracy. Secondly, the numerical product is needed in the implementation of quasi-synchronous sampling algorithm. Based on the derivation process of quasi-synchronous sampling algorithm based on complex rectangle, the improved quasi-synchronous sampling frequency measurement algorithm is studied and deduced, including the quasi-synchronous sampling frequency measurement algorithm based on complex trapezoid and the quasi-synchronous sampling frequency measurement algorithm based on complex Simpson. The improved quasi-synchronous sampling power system is established. The computational complexity of the commonly used windowed interpolation FFT algorithm and the improved quasi-synchronous sampling frequency measurement algorithm is theoretically compared and analyzed. The results show that the number of additions needed in the implementation of the improved quasi-synchronous sampling frequency measurement algorithm is much less than that of the addition algorithm when the sampling frequency and the sampling points are the same. Because the complex rectangle, complex trapezoid and complex Simpson are only different in integral weighting coefficients, and the computational complexity of the quasi-synchronous sampling frequency measurement algorithm is only related to the number of sampling points, the computational complexity of the two improved quasi-synchronous sampling frequency measurement algorithms studied in this paper is phase. Similarly, the accuracy of the improved quasi-synchronous sampling frequency measurement algorithm and the windowed interpolation FFT algorithm are compared and analyzed through simulation experiments. In this paper, the interpolation FFT algorithm based on 4-term 3-order Nuttall window is selected for comparison, and the fundamental frequency fluctuation, harmonic interference and white noise are simulated and analyzed. The simulation results show that the improved quasi-synchronous sampling frequency measurement algorithm has higher accuracy than the interpolation FFT algorithm based on 4-term 3-order Nuttall window, and the algorithm can effectively suppress the influence of white noise on power system frequency measurement. The performance of the quasi-synchronous sampling frequency measurement algorithm based on complex trapezoid is slightly better than that of the quasi-synchronous sampling frequency measurement algorithm based on complex Simpson. Finally, the improved quasi-synchronous sampling frequency measurement algorithm proposed in this paper is implemented by using embedded system platform, and according to GB/T 15945-2008, GB/T 19862-2005, DL/T 1028-2005. According to the requirement of 2006, a frequency measurement and calibration scheme for power system is designed. Through a large number of frequency measurement experiments before and after calibration, the power system frequency measurement deviations before and after calibration are compared and analyzed, and the uncertainty of power system frequency measurement after calibration is analyzed when the fundamental frequency is invariant. It can meet the requirement of the national standard that the frequency deviation is within (+0.01 Hz), which also verifies the effectiveness and accuracy of the improved quasi-synchronous sampling frequency measurement algorithm in practical application.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM935.1

【参考文献】