边缘检测在HVDC系统故障诊断中的应用  

Wavelet Edge Detection Based HVDC System Fault Identification

作　　者：高冬梅[1] 袁海文[1] 张永斌[1] 曾倩[1] 

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100083

出　　处：《高电压技术》2008年第11期2380-2384,共5页High Voltage Engineering

摘　　要：为了及时分辨HVDC系统故障类型并快速恢复,对直流输电标准测试系统在各种故障,包括交流系统发生故障、逆变换相失败故障以及直流线路故障下的典型响应特性进行了分析。采用劋trous小波方法将行波信号分解成不同尺度下的小波面,其中包含了该尺度下细节信息,实现对故障的边缘检测;提出了3种故障判别准则,推出了模极大值能量的概念,并利用反射行波信号的小波变换模极大值的幅值、极性及模极大值能量等变化规律对不同HVDC系统故障进行诊断,实现逆变器换相失败和交流系统单相故障的识别。MATLAB仿真结果表明提出的方法能很好地完成HVDC系统的故障诊断。The detection and fast recovery of faults in HVDC（ High Voltage Direct Current） system are important for a safe operation of power system. It is not easy to identify HVDC power system faults by using pure frequency do main based methods or pure time domain based methods. Based on wavelet theory the sudden change can be detected and identified between different transients with help of wavelet modulus maxima. This paper describes a new highspeed HVDC power system fault identification based wavelet edge detection technique, using the representation of the traveling waves through wavelet modulus maxima, criterions for similar transient signals such as HVDC line fault, commutation failure in the converter station and AC single phase short circuit fault at AC side, are proposed. In this paper, the behaviour of different faults in HVDC power system is analysed through wavelet transform. Considering a better time location and a better information keeping, the paper use the trous structure but without down sampling block after high pass filters. The trous decomposition is one of discrete wavelet transform algorithms. On the basis of trous wavelet decomposition theory a detecting edge method is derived, according to which the signal can be decomposed into wavelet planes of increasing scales. Based on trous algorithm the fault detedtion result is satisfactory. In this paper three criterions for identification are proposed. MATLAB simulation results show that these criterions based on wavelet modulus maxima can make a definite identification of HVDC system faults, commutation failure and AC single phase fault. The application of wavelet edge detection technique leads to a faster, easier and more reliable protection solution and development of new fast identification technique for HVDC power system.

关 键 词：故障诊断 反射行波信号 边缘检测 小波变换 模极大值 换向失败 dtrous小波分解 

分 类 号：TM721[电气工程—电力系统及自动化]