小波软阈值去噪技术在电能质量检测中的应用

小波方法是一种很好的电能质量信号检测和分析工具，但其性能往往受信号中噪声的影响，当噪声比较大的时候，小波方法甚至会失效。文中根据小波变换的时频特性，分析了信号和噪声在小波分解过程中的不同特性，并在此基础上利用改进的软阈值去噪技术对电能质量信号进行信号去噪处理。软阈值方法能根据各小波空间上特征分量和噪声的统计特性设置适当的阈值来消除噪声，并以此恢复小波方法的性能。该方法不仅较好地解决了保护信号局部特征与抑制噪声之间的矛盾，能很好地对各种电能质量信号进行去噪处理，而且达到了数据压缩的效果。仿真计算结果表明，该去噪方法是有效的。

PPLICATION OF WAVELET SOFT-THRESHOLD DE-NOISING TECHNIQUE TO POWER QUALITY DETECTION

The wavelet transform (WT) technique is good for detecting, localizing and analyzing power quality disturbance signals. However, as the signals under analysis are always disturbed by noises, the effect of the WT technique is degraded. Strong noises even lead to a wrong judgement. This paper first analyses the different characteristics of signal and noise in wavelet analysis. Then an improved soft threshold algorithm is proposed to enhance the WT technique in processing the noise riding signals. With this algorithm, a proper threshold is formed according to the signal's eigenvalue and the statistic characteristic of noise in each wavelet spaces to de noise the signals under investigation. Hence, the performance of the WT in detecting and analyzing the signals can be restored. This method not only can solve the conflict commendably between the local characteristics of protection signal and restriction of noises, thus properly de noise the quality signals, but also realizes the compression of the data. Finally, the validity of the proposed method is verified by the results of the simulation.