摘要
在小波变换数据压缩方法和神经网络数据压缩技术的基础上,提出了将小波和神经网络应用于电能质量扰动信号数据压缩的方法。利用小波时域和频域的双重分辨率和神经网络的非线性函数逼近能力,以压缩比、均方误差为压缩效果的评价指标,对实际扰动信号进行数据压缩。采用样条小波和径向基神经网络数据压缩方法,以一个实例,给出了电能质量扰动信号的压缩仿真过程,给出了各类(电压凹陷、突起、尖峰、闪变及瞬态振荡)电能质量扰动信号的仿真分析结果。结果表明,该电能质量扰动信号数据压缩方法,压缩后得到的均方误差为-16.1397 dB,压缩效果良好。
A method combining data compression technologies using wavelet transform and neural network is presented to compress the data of power quality disturbance. It uses double resolutions of wavelet in time & frequency domains and the nonlinear fitting ability of neural network to compress practical disturbance signals,taking the compression rate and root mean squared error as evaluation indices. The spline wavelet and the radial basis function neural network are adopted. The compression process is simulated,and results of different disturbances (voltage sag,swell,peek,flicker and transient surge) are provided. The root mean squared error after compression is -16.139 7 dB,showing its effectiveness.
出处
《电力自动化设备》
EI
CSCD
北大核心
2007年第3期38-40,56,共4页
Electric Power Automation Equipment
关键词
电能质量
数据压缩
小波变换
神经网络
power quality
data compression
wavelet transform
neural network