摘要
准确识别故障类型是实现配电网故障定位的前提。应用小波变换技术提取反映接地故障特征的零序电压低频信号能量,应用小波变换和信息熵相结合的方法提取三相电压的小波奇异熵。以零序电压低频能量和三相电压的小波奇异熵为输入特征量,以相别A、B、C和地G为输出量,建立了四输入四输出的SVM故障类型识别网络。应用ATP/EMTP搭建配电网仿真模型模拟了各种故障条件下的各种故障类型。仿真分析表明,该方法能够快速准确地识别各种故障类型,且不受过渡电阻、故障位置等的影响。
The accurate identification of fault types is the foundation of realizing the location of grid fault. The characteristics of ground fault zero sequence voltage low-frequency signal energy is obtained by application of wavelet technology. The wavelet singular entropy of three-phase voltage is extracted by the method of combining wavelet transform and information entropy. Taking the low-frequency energy of zero sequence voltage and wavelet singularity entropy of three-phase voltage as the characteristic input and taking phase type A, B, C and G as output, the four-input and four-output of the Support Vector Machine (SVM) network is formed for fault type recognition. The distribution network simulation model is built by application of ATP/EMTP to simulate various fault types under various fault conditions. Simulation and analysis show that this method can identify the fault types rapidly and accurately and it won't be affected by transition resistance or faulty location and so on.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2011年第23期16-20,共5页
Power System Protection and Control
基金
国家自然科学基金资助项目(60971077)~~
关键词
配电网
小波变换
小波奇异熵
故障类型识别
支持向量机
distribution network
wavelet transform
wavelet singularity entropy
fault type recognition
support vector machine