基于小波变换能量分布和神经网络的电能质量扰动分类

提出了基于小波变换能量分布和BP神经网络的电能质量扰动的自动分类方法.利用小波变换对电能质量扰动信号进行多分辨分析,计算各分解层能量分布,求出该能量分布与标准信号能量分布差值并将其作为信号特征量,通过一个3层BP网络得到扰动的类型.该方法将小波变换系数转化为能量分布,减少信号特征的数量,从而简化了神经网络结构.测试结果表明,即使在较强噪声信号背景下,该方法对电能质量扰动类型的识别率仍可达到94.5%,证明了该方法的有效性.

Classification of power quality disturbance based on wavelet energy distribution and neural network

An automatic method of PQ(Power Quality) disturbance classification is proposed,which uses wavelettrans form to perform multi-resolution analysis of the original PQ signals and calculates the energy distribution of each level.The difference of energy distribution between the original PQ signals and the standard signals is used as the input of a three-layer BP neural network and its output is the classification of PQ disturbance.The method transforms wavelet coefficients into energy distribution to reduce th...