基于双曲S变换的变压器励磁涌流和内部故障识别新方法  

A new method of identifying inrush and internal faults of power transformer based on H-S transform

作　　者：焦尚彬[1] 黄璜[1] 赵黎明 张青[1] 

机构地区：[1]西安理工大学信息与控制工程研究中心,陕西西安710048 [2]许继电气公司,河南许昌461000

出　　处：《电力系统保护与控制》2011年第16期114-120,138,共8页Power System Protection and Control

基　　金：陕西省自然科学基金资助项目(2009JQ7009);陕西省教育厅专项基金资助项目(2010JK750);西安理工大学科技创新计划

摘　　要：提出了一种基于双曲S变换识别励磁涌流和变压器内部故障的新方法。利用S变换具有优秀时频呈现能力和时频分辨率的特点,采用双曲窗函数,把励磁涌流和故障电流信号变换到相空间进行特征提取,从经过变换以后得到的时频等值线图可以发现二者具有明显不同的特征。通过计算H-S变换后的能量以及1-9层能量与1层能量的比值,发现比值6是区别励磁涌流和故障电流的阈值。仿真分析结果和动模试验均表明:H-S变换可以更清楚地反映信号的时频局部化特性,能准确识别励磁涌流;并且具有较好的抗干扰能力,在变换域易于信号和噪声的分离。A new method to identify inrush and internal fault of transformer based on hyperbolic S-transform （HST） is discussed. This paper uses the S transform （ST） which has excellent characteristics of time-frequency skills and time-frequency resolution and adopts the Hyperbolic window function. The signals of inrush current and fault current are transformed to phase space by using HST to extract the feature, the time-frequency contours are obtained. It is found that the time-frequency contours in case of inrush are different from that in case of internal faults. By calculating the energy after H-S transform and the ratio of energy level 1-9 and level 1 we found that ratio 6 is the threshold between inrush and fault current. The simulation result and dynamic test show that H-S transform can reflect the signal＇s time-frequency localization clearly, can identify inrush current, and it has strong capability in noise reduction, and is easy to separate signal from noise in ST domain. This work is supported by Natural Science Foundation of Shaanxi Province （No. 2009JQ7009）.

关 键 词：变压器差动保护 励磁涌流 内部故障 H-S变换 信噪分离 

分 类 号：TM407[电气工程—电器]