誠成设计,改进的多小波去噪算法在XLPE电力电缆中的应用.docVIP

改进的多小波去噪算法在XLPE电力电缆中的应用 摘要：在电能质量检测应用中，电网谐波由两部分组成，一部分为整数倍基波频率，另一部分为非整数倍基波频率（即间谐波），间谐波分布在除基波和谐波以外的整个频谱范围内，具有不确定性和非平稳特性。目前间谐波检测方法主要有：改进的FFT算法、小波分析、小波包分析、神经网络、prony分析等[1]，采用FFT算法选择适当的采样频率和频率窗周期性同步采样，难以精确地测得间谐波。其中，prony算法可以精确地测得间谐波成分，但是其对噪声比较敏感。而实际电网信号总是存在噪声，所以在检测之前，首先要对采集的谐波进行消噪预处理。多小波保持了单小波所具有的良好的时域与频域的局部化特性，又克服了单小波的缺陷，多小波可同时具有光滑性、对称性、正交性、短支撑性，高阶消失矩等属性。本文对小波软、硬阈值去噪方法进行了仿真，根据Matlab仿真结果，针对小波软、硬阈值方法的缺点，提出改进的多小波阈值去噪算法，对XLPE电力电缆中信号进行多小波分解后再去噪预处理，其去噪效果得到了显著提高。同时就全局处理采用统一的阈值对各级高频系数进行滤波，由于噪声对应的小波系数随着分解尺度的增加，其幅值会减小，所以统一阈值消噪会有较大的误差的缺点，采用分层（分频带）处理。针对各分解层采用不同的阈值进行滤波，分层阈值估计相应调整各层阈值，以达到更好的去噪效果。对比仿真结果发现，改进后的多小波阈值去噪取得了更好的去噪效果。 关键词：多小波，小波去噪，电网谐波，分频带，分层阈值 Abstract：In power quality testing applications, grid harmonics consists of two parts,one part is an integer multiple of the fundamental frequency, another part is the non-integer multiple of the fundamental frequency (ie,interval harmonics).The entire spectral range inter-harmonic distribution in addition to the fundamental and harmonic and with the uncertainty and non-stationary characteristics. It is difficult to accurately measured the harmonics using FFT algorithm to select the appropriate sampling frequency and the frequency window of periodic synchronous sampling. Currently , the detection methods the interval harmonic and harmonic: Improved FFT algorithm, wavelet analysis, wavelet packet analysis, neural network, prony analysis, etc. Which, prony algorithms can accurately measure the harmonic components ,but it is more sensitive to noise. The actual grid signal there is always noise, denoising preprocessing shouldbe taken before, we must first harmonic collected. Multiwavelet maintain the characteristics that single wavelet has a good time and frequency domains of the local, but also overcome the defects of the single wavelet, multiwavelet can be smooth, symmetry, orthogonality, short support, high-enddisappear and other properties. The wavelet soft and hard threshold denoising method are simulated, Matlab simulation results