摘要
根据奈奎斯特采样定理,对电网使用传统的时域采样方法进行采样将会产生海量数据,这些数据要经过前端压缩处理才能够传输会后台。而近年来新提出的压缩感知理论指出,对于满足一定稀疏性条件的信号,后台可以通过少量的数据就可以恢复原信号,当稀疏性越好,信号恢复效果越好。文章针对电网电压的特点,将离散傅立叶变换和多尺度小波变换结合起来,构造出一种新型的稀疏矩阵,以提高电网受扰动时采样信号的稀疏性。Matlab仿真结果显示,新构造的稀疏矩阵,在面对短时高频的电能质量信号干扰时,比常规使用的离散傅立叶变换矩阵有更好的恢复效果。
Traditional method of measuring the power grid data causes huge amount of data due to the Nyquist sampling theorem, and those data must be compressed in order to transmit them to the back end. Compressed sensing theory that was proposed in the recent years, suggests that sparse signal can be accurately recovered through only a few amount of data. In the paper, we create a new type of sparse matrix based on Discrete Fourier Transform and multiresolution Wavelet Transform, in order to increase the sparsity of the disturbed power grid signal. Matlab simulation shows that this new type of sparse matrix is more adaptable to different types of disturbed power grid signal than the traditional Discrete Fourier Transform matrix.
出处
《电子技术（上海）》
2017年第10期61-64,共4页
Electronic Technology
关键词
电能质量
压缩感知
稀疏矩阵
小波变换
信号处理
electrical power quality
compressed sensing
sparse matrix
Wavelet transform
signal processing
