利用小波包变换实现噪声环境下特征信号的提取

电力系统暂态或故障时 ,电压电流信号包含了故障的信息 ,文中称包含了故障或暂态信息的电压和电流信号为特征信号 ,利用特征信号可以进行系统分析和故障检测 ,但是特征信号往往被淹没在大量的噪声信号中 ,这样给电力系统分析和检测带来困难。文中分析了电力系统中几种常见的噪声和特征信号的时频特性 ,简单介绍了小波包变换的理论和特点 ,分析了利用小波包变换来消除噪声 ,提取特征信号的理论 ,并通过实例验证了小波包良好的抗噪能力 ,为实现噪声环境下特征信号的提取提供了良好的分析方法 ,为电力系统分析和故障检测提供了良好的工具。

Character signal extraction under noise background with wavelet packet transform

The voltage and current signals in power system often contain the character of the fault or transient when there is a fault or a transient in the power system. In this paper the voltage or current signal containing the fault or transient information is called character signal, which can be used in system analysis and fault detection. But these character signals often submerged in noise signals. The paper analysis the time-frequency character of noise and the character signals, give a brief summary of the theory of wavelet packet transform, thus analyze the theory bases of the wavelet packet based method of character signals extraction under noise environment. The simulation validates the good performance of wavelet packet transform in signal and noise separation, thus present a good tool for the wide use of wavelet packet transform in power system analysis.