广域测量系统数据在线无损压缩算法

广域测量系统(wide-area measurement system，WAMS)中，相量测量装置(phasor measurement unit，PMU)上传数据频率为100Hz，存储占用空间和写入速度对系统要求很高。旋转门(swing door trending，SDT)压缩算法在保留原始数据足够精度的条件下，压缩效率低，Huffman编码压缩计算时间长，难以适应WAMS实时数据的采样速度。文中给出了一种WAMS实时数据在线压缩算法：对原始数据做初等变换，只保存过程数据的增量，浮点(float)型数据分解为单字节存取，用改进的LZW(Lemple-Ziv-Welch)算法实现WAMS过程数据的在线无损压缩。6 000个连续实时数据压缩计算时间为16 ms，压缩比小于30%，可以满足电力系统分析计算对过程数据的要求。工程实践表明，该算法有效可靠，可满足WAMS实时数据的存储要求。

Online Lossless Compression of WAMS Data and Its Algorithm Implementation

In wide-area measurement system (WAMS) the data uploading frequency of phasor measurement units (PMU) is 100 Hz,to meet the high requirement to occupied storage space and writing speed of this system,the PMU data should be compressed. To reserve enough accuracy of raw data,the data compression efficiency of swing door trending (SDT) algorithm is too low; and the calculation time of Huffman coding compression algorithm is too long,thus it is difficult for these compression algorithms to adapt to the sampling speed of real-time data in WAMS. In this paper an online compression algorithm for real-time WAMS data is proposed,in which the elementary transformation of raw data is performed and only the increments of process data are reserved; the float data is decomposed into single bytes to be accessed; and the online lossless compression of process data in WAMS is realized by improved Lemple-Ziv-Welch (LZW) algorithm. Using this data compression method,the compression calculation time for 6000 continuous real-time data is 16ms and the compression ratio is lower than 30%,therefore the demand of power system analysis and calculation on process data can be satisfied. Engineering experiences show that the proposed data compression algorithm is effective and reliable,and can meet the requirement of real-time data storage in WAMS.