基于小波变换的数字滤波算法

Digital Filtering Arithmetic Based on Wavelet Transform

在微机保护装置中,数字滤波是非常重要的环节,尤其在故障暂态当中,信号通常含有大量的谐波,但此时采用间断频谱的傅氏级数分析方法已不合适。而将小波分析技术应用于信号的数字滤波以及算法分析,可实现谐波分量的有效分离。为此,基于紧支撑正交小波变换理论和多分辨率分析特性的应用研究,提出了信号的数字滤波算法方案,并通过仿真对算法进行了验证和比较,仿真分析表明,基于小波变换的滤波方案能够有效地滤除衰减直流分量、噪声、频率波动和非整次谐波的影响。

Digital filtering is an extremely important link in microcomputer-based protection system. Especially in transient state of a fault, when the signal usually contains volumes of harmonics, it is not suitable to adopt Fourier series analysis of discontinuous spectrum, while the application of wavelet analysis to signal digital filtering and arithmetic analysis can effectively separate harmonic component. Based on application study of close support cross wavelet transform theory and its multi-resolution analysis characteristic, the digital filtering arithmetic for signal is put forward. The arithmetic is validated and analyzed by simulation. The results prove that the digital filtering arithmetic based on wavelet transform is capable of filtering direct current component, noise, freouency jitter and non-integer order harmonic efficiently.