暂态电能质量扰动信号检测方法的研究

发布时间：2018-06-27 03:11

本文选题：暂态电能质量 + 扰动检测　；参考：《东北石油大学》2017年硕士论文

【摘要】：近年来,电能质量问题引起了电力部门和用户的高度重视,在现代社会中已逐渐成为使用最广泛且不可或缺的一种能源。由于大量精密仪器和电力电子设备的广泛使用,使得电网电能质量日益恶化,必可避免的将造成电能质量的下降。分析研究电能质量扰动信号,已渐渐变成现代电力系统的首要研究课题,具有重要的探究意义。首先,本文归纳了国内外在电能质量检测方面所做的研究,对电能质量的概念以及分类方式进行不同角度的描述,剖析并总结了电能质量扰动信号的特征,并给出了扰动信号的数学模型以及对应的波形图。深入分析傅立叶变换、短时傅立叶变换以及小波变换的基本原理,探究并对比了其在电能质量分析范畴的应用。其次,本文考虑到在实际的工程应用中,由于外界因素的干扰,检测到的信号通常包含噪声。在采用小波去噪时,由于存在少许的区域,其中的原信号与噪声信号相对更靠近,容易错误的将原信号的部分细节看做是噪声信号,从而给去除掉。为了更好的保留原始信号的有用信息,本文提出一种基于系数变换的改进的去噪算法。通过MATLAB仿真,证实了该算法能够在去除噪声的同时,更好的重现原信号。再次,由于小波变换拥有很好的时频局部化特性,同时依据暂态电能质量信号的不平稳特征,则信号的奇异性能够利用小波变换模极大值来体现。本文在小波变换的前提下,提出一种动态窗模极大值扰动定位算法,针对传统模极大值在检测信号突变点时存在的突变点遗漏和过零点检测不明显的问题进行算法改进,通过对单一和复合扰动的仿真实验,证明了本文所提方法的可行性。最后,运用S变换法对这些扰动信号进行检测分析,从S变换结果矩阵中我们能够提取出扰动信号起止时间、信号的扰动幅值和频率特征量,绘制S变换模矩阵列向量平方和均值曲线、列向量极大值曲线和行向量平方和均值曲线,通过三维网络图和三种曲线,可以清晰看到信号的扰动特征,并检测出信号的扰动时间、幅值和频率,证实了S变换在暂态电能质量扰动检测方面的优势。
[Abstract]:In recent years, the power quality problem has attracted the attention of the power department and the users, and has gradually become the most widely used and indispensable energy in modern society. Because of the extensive use of a large number of precision instruments and power electronic equipment, the power quality of the power grid is becoming worse and worse, and the quality of power can be avoided. The analysis and study of power quality disturbance signal has gradually become the primary research topic of modern power system, which has important research significance. Firstly, this paper summarizes the research on power quality detection at home and abroad, describes the concept and classification of power quality in different angles, and analyzes and summarizes the power quality disturbance. The mathematical model of the disturbance signal and the corresponding waveform diagram are given. The basic principles of Fu Liye transform, short time Fu Liye transform and wavelet transform are deeply analyzed, and their application in the field of energy quality analysis is explored and compared. Secondly, this paper considers that the external factors are dry in practical engineering applications. The signal detected usually contains noise. In the use of wavelet de-noising, the original signal is relatively close to the noise signal due to the existence of a few regions, and it is easy to mistake the partial details of the original signal as a noise signal. The improved denoising algorithm of coefficient transformation. Through MATLAB simulation, it is proved that the algorithm can better reproduce the original signal while removing the noise. Thirdly, the wavelet transform has a good time-frequency localization characteristic, and the singularity of the signal can make use of the wavelet transform mode pole according to the unstable characteristic of the transient power quality signal. In this paper, a dynamic window mode maximum disturbance localization algorithm is proposed on the premise of wavelet transform. The algorithm improves the problem of the missing point mutation point and the zero crossing point detection of the traditional mode maximum value when detecting the signal mutation point. Finally, the S transform method is used to detect and analyze the disturbance signals. From the result matrix of the S transformation, we can extract the time of the disturbance signal, the amplitude and the frequency characteristic of the signal, and draw the vector square and mean curve of the S transform matrix array, the maximum curve of the column vector and the mean square and the mean of the row vector. The curve, through the three-dimensional network graph and the three kinds of curves, can clearly see the disturbance characteristic of the signal, and detect the disturbance time, amplitude and frequency of the signal, and confirm the advantage of S transform in the detection of transient power quality disturbance.
【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM930

【参考文献】