基于B样条小波的电力系统故障录波信号的数据压缩技术  

B-Spline Wavelet Based Data Compression Technology for Power System Fault Recorder

作　　者：丁泽京[1] 谢胜曙[1] 曹志瑞[1] 李孟秋 

机构地区：[1]湖南大学电气与信息工程学院,410082 [2]湖南安力电子公司,410082

出　　处：《电气应用》2006年第8期31-34,共4页Electrotechnical Application

摘　　要：针对在电力系统发生故障时,微机型继电保护装置因受内存容量的限制,不能存储过多故障数据的现象,提出一种实用的数据压缩技术———采用离散小波变换把数据从时域变换到频域,对高频系数选取门限进行阈值量化,对低频系数则经过数据变换后实施自适应Huffman编码处理,并对小波阈值设置、样条小波、Huffman编码等关键问题进行了详细讨论。仿真分析表明,数据压缩比和重构的方均误差均很小,验证了本方法的可行性,具较高的实用价值。Due to the restriction of limited memory, microcomputer relay protection can not store more data when power system is in fault. A practical data compression technology is proposed. Data are firstly transformed from time domain to frequency domain by DWT. The higher frequency coefficients are quantized by adaptive threshold, and then the lower frequency coefficients are compressed by adaptive Huffman coding. Key issues such as the choice of threshold, B-spline wavelet and adaptive Huffman coding are discussed in detail. Experimental results demonstrate that both data compression ratio and reconstruction error are very small , which verifies the proposed algorithm is feasible and of great practical value.

关 键 词：离散小波变换 自适应 HUFFMAN编码 门限阈值 数据压缩 

分 类 号：TM711[电气工程—电力系统及自动化]