基于压缩感知的电能质量数据压缩技术研究  

Research on Power Quality Data Compression Technology Based on Compressed Sensing

作　　者：纪萍[1] 陈玲[1] 胡娟[1] 

机构地区：[1]河海大学文天学院电气信息工程系,安徽马鞍山243000

出　　处：《控制工程》2016年第11期1747-1751,共5页Control Engineering of China

基　　金：河海大学文天学院校级科研项目(WT14002)

摘　　要：电能质量数据采集具有采样频率高,监测时间长的特点。有效的对电能质量数据进行压缩是非常有必要的。现有的数据采样压缩方法,是基于奈奎斯特采样定理,具有采集数据量大,压缩效果差的缺点。文中引用新的压缩方法-压缩感知理论,突破传统的采用定理的限制,将采样和压缩结合在一起,大幅降低了数据采样量。该方法针对电能质量数据特点,采用傅里叶稀疏基和OMP重构算法进行数据压缩重构。通过与小波变换技术进行仿真对比显示,其不仅能有效对数据进行压缩,并且数据压缩的性能指标有了很大的提高,为电力系统数据压缩提供了新的分析处理方法。The power quality data collection has the characteristics of high sampling frequency and long monitoring time. It is necessary to compress the power quality data. The existing data sampling and compression method, based on Nyquist sampling theorem, has the disadvantages of large collection data, poor compression effect. In this paper, a new compression method - compressed sensing, which breaks through the limitation of the traditional theory, is used to combine the sampling and compression. In view of the signal characteristic of the power system, the method uses Fourier sparse and OMP algorithm for data compression. Compared with the wavelet transform technology, the method can not only compress the data effectively, but also improve the performance of data compression. It provides a new analysis method for power system data compression.

关 键 词：压缩感知 小波变换 稀疏基 正交匹配追踪算法 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]