基于拟合提升复小波的电能质量暂态事件分析研究

Stable and excellent power quality provides basic guarantee for the economic development. While realizing accurate detection of disturbances that occurred in power supply is of great importance for power quality improvement. In this subject, aiming at the transient problems of power quality, the analysis on complex wavelet construction based on fitting and lifting as well as implementation of the corresponding transforms are supposed to be investigated and then applied to disturbance detection of the transient events. According to the theory of complex wavelet, the lifting algorithm based on fitting will be adopted for complex wavelet construction, and the relations between parameter selection and the characteristics of the new constructed complex wavelet will be studied. For signal’s perfect reconstruction, the realization for lifting complex wavelet decomposition and reconstruction on the basis of predict and update steps will be analyzed, and then the single branch reconstruction and redundant algorithm will also be discussed, due to which basic theory of the fitting based lifting complex transforms will finally be formed. In view of disturbance detection for the power quality problem, the method for singularity component detection based on fitting lifting complex wavelet will be explored to effectively remove noise and successfully extract the composite feature information and how the amplitude-frequency characteristic and phase-frequency characteristic of complex wavelet will impact the results is also going to be discussed, with which the disturbance can be well detected so as to offer efficient theory support for reliable power supply.

稳定优质的电能质量是国民经济快速发展的基本保证。实现对供电中扰动的准确检测对于提高电能质量具有极其重要的意义。本课题以电能质量中的暂态问题为主要研究对象，拟对基于拟合提升的复小波构造和变换方法进行研究，并将其应用于暂态事件的扰动检测研究。根据复小波的基本理论，拟采用基于拟合的提升算法来实现复小波的构造，分析不同参数的选取与新生成复小波特性之间的相互关系。以信号的完全重构为依据，研究基于预测和更新步骤的提升复小波变换分解和重构的具体实现方法，并在此基础上结合信号特征提取的需要，研究单支重构和冗余算法下的变换实现方法，最终形成基于拟合提升的复小波变换的基本理论基础。针对电能质量中的扰动检测问题，研究基于拟合提升复小波分析的奇异性成分检测方法，有效滤除背景噪声，充分提取复合特征信息，探讨复小波不同的幅频特性和相频特性对分析结果的影响，实现对扰动的准确检测，为可靠供电提供有效的理论支持。

稳定优质的电能质量是国民经济快速发展的基本保证。实现对供电中扰动的准确检测对于提高电能质量具有极其重要的意义。. 本课题以电能质量问题暂态事件扰动的准确检测为主要研究目标，对基于拟合提升的复小波构造和变换方法进行了研究。. 根据复小波和提升算法的基本理论，对对偶提升算子/提升算子进行了复数设计，推导得到了基于复数算子实部和虚部独立接续提升的双步提升方案，用以对初始双正交滤波器组进行复提升而获得具有新特性的复数双正交滤波器组。在此基础之上，结合提升小波变换可完全在时域进行从而保证信号完全重构的重要特性，提出了基于双树结构的提升复小波变换正向分解和逆向重构的具体实现过程。同时，针对信号特征提取的需要，研究了单支重构和冗余算法下的提升复小波变换实现方法，最终形成了基于拟合提升的复小波变换的基本理论基础。. 根据数据拟合和插值细分的相关理论，对基于拟合和提升算法的复小波构造方法进行了研究。分析了不同参数的选取与新生成复小波特性之间的相互关系。. 为实现对电能质量暂态事件的扰动的更精确识别，探讨了分析过程中重要参数对于扰动起止时刻定位的影响。提出了基于邻域相关系数的提升小波包降噪方法，用以有效滤除背景噪声。探讨了提升小波包中的最优节点选取问题和基于能量分析的扰动分类识别方法。. 本课题研究提出的提升复小波的基础理论，为电能质量暂态事件扰动的准确检测和高精度定位提供了行之有效的分析方法，也为用户端的可靠供电提供了理论依据和技术支持。