基于FRFT的电能质量扰动滤波算法研究

发布时间：2018-07-15 08:43

【摘要】：随着电子设备的大量投入以及较大负荷的危害,电网的电能质量逐渐下降,电能质量问题受到各行各业和消费者的越来越多的关注,对扰动信号正确的滤波与检测是改善和提高电能质量的一个重要前提。产生电能质量问题的扰动是多种多样的,可以分为稳态和暂态两类。对于少数的扰动可以使用时域或频域分析,当前主要采用的是时频联合的方式对电能质量问题进行分析。本课题主要研究的是暂态电能质量扰动信号,为提高电能质量,我们应选择恰当的方法对待测信号进行滤波,然后才能对各种暂态问题做较好的检测、评估及分类。根据暂态电能质量的非平稳性,本文对几种暂态信号滤波方法做了描述,基于快速傅里叶变换(FFT)的滤波方法是应用最为广泛的一种算法,但它只适用于分析平稳信号,对于非平稳信号,本文采用的是基于分数阶傅里叶变换(Fraction fouriertranform,FRFT)的暂态电能质量扰动滤波。FRFT虽然对于暂态信号的处理效果较好,但无法表征信号局部特征,而传统小波在时频域内分析不足,且在某种情况下对噪声不敏感,为了弥补这些缺点,在FRFT的基础上研究分析了分数阶小波变换。利用MATLAB/Simulink的仿真平台搭建模型,仿真输出了 FRFT对加噪信号的滤波结果,对比分析实验数据,表明FRFT的滤波效果比FFT的好,证实了 FRFT滤波的有效性;且分数阶小波变换滤波的计算量少,滤波效果好,能够较好的还原波形及其特性,克服了单一算法的缺点和不足,验证了滤波算法的可行性。本文的滤波方法对于暂态信号有较好的滤波效果,可对暂态扰动信号做有效分析。
[Abstract]:With the large input of electronic equipment and the harm of large load, the power quality of the power grid is gradually decreasing. The problem of power quality has attracted more and more attention from all walks of life and consumers. The correct filtering and detection of the disturbance signal is a important prerequisite for improving and improving the power quality. It can be divided into two classes of steady state and transient state. For a few disturbances, the time domain or frequency domain analysis can be used. The current mainly uses the time frequency combination to analyze the power quality problem. The main research is the transient power quality disturbance signal. In order to improve the power quality, we should choose the appropriate method to treat it. The signal is filtered, and then a variety of transient problems can be detected, evaluated and classified. According to the non-stationary of the transient power quality, this paper describes several transient signal filtering methods. Based on the fast Fourier transform (FFT) filtering method, it is the most widely used algorithm, but it is only suitable for the analysis of stationary signals. For non-stationary signals, the transient power quality disturbance filter based on fractional Fourier transform (Fraction fouriertranform, FRFT) is used in this paper. Although the transient signal processing effect is better, it can not characterize the local characteristics of the signal, but the traditional wavelet is not enough in the time-frequency domain, and in some cases the noise is not sound. In order to make up these shortcomings, the fractional wavelet transform is studied and analyzed on the basis of FRFT. The simulation platform of MATLAB/Simulink is used to build the model, and the result of the filtering of FRFT to the noise signal is simulated. The comparison and analysis of the experimental data shows that the filtering effect of the FRFT is better than that of the FFT, and the validity of the FRFT filter is confirmed and the fractional order is small. The wave transform filter has less computation, better filtering effect, better reduction waveform and its characteristics, overcomes the shortcomings and shortcomings of the single algorithm, and verifies the feasibility of the filtering algorithm. The filtering method of this paper has good filtering effect on transient signals and can effectively analyze transient disturbance signals.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711

【相似文献】