基于小波变换的低压电力网载波通信研究

WAVELET TRANSFORM BASED LOW VOLTAGE POWER LINE TELECOMMUNICATION

小波分析是一种优于传统信号分析方法的时频分析方法。由于同时具有时域和频域的良好局部化特性及自动调节时频窗的特点 ,可以聚焦到被分析信号的任意局部细节 ,具有分析效率高、质量好的优点 ,因此近年来被广泛用于信号处理。低压电力网载波通信也是近年来电力系统中的一个热门课题 ,具有许多优点 ,然而由于低压电网噪声成分复杂 ,随负荷变化大 ,具有很强的非平稳特性 ,因此又是一个极具挑战性的课题。文章探索了一种能对低压电网通信信号进行有效处理的方法 ,采用取不同 N值的 Daubechies小波对实际电力系统中采集的几种载波通信信号进行分析计算 ,得出的结论为 :Daubechies小波可以对通过低压电力网进行通信的载波信号进行有效的处理。采用不同 N值的 Daubechies小波可以得到不同的处理效果 ,滤波器的长度越长其滤波性能越好 ,但信号处理的延时也越长 ,因此实际应用时必须综合考虑滤波性能、时延及实际需求等多方面的因素 。

As a useful time frequency signal analysis method better than traditional ones for signal analysis, wavelet analysis possesses good localization characteristic in time domain and frequency domain and can self adjust the time domain window. So it can be focalized to any partial detail of analyzed signal. With this method the analysis results and efficiency are satisfactory. It is being widely used in signal processing. As a topic receiving special attention in recent years, low voltage power line telecommunication (PLT) has many advantages. However, the noise in low voltage power network has complicated components and varies severely with the variation of power load, so that the PLT signal transmitted in low voltage network may be seriously interfered. In this paper a method which can effectively process the communication signals in low voltage power line is explored, in which the Daubechies wavelet with different value of N is adopted to calculate and analyze several PLT signals acquired from practical power network under different signal transmission conditions, and the following conclusions are obtained: (1) Daubechies wavelet can process PLT signal in low voltage power network effectively. (2) Using Daubechies wavelet with different values of N, different processing results are obtained. (3) The longer the length of the filter is, the better filtration result will be, but the time for signal processing will be longer correspondingly. (4) In practical application, the factors such as filtration performance, time delay and practical requirements, etc., should be comprehensively considered for selecting appropriate wavelet function.