摘要
为了识别输电线路单相接地瞬时性和永久性故障,利用经验模态分解将电压暂态信号分解为若干个固有模态函数,选取代表信号较高频部分的前四层固有模态函数进行归一化,作为最小二乘支持向量机的输入特征量。大量EMTP仿真和Matlab计算表明该方法能很好地区分瞬时性故障和永久性故障,表明该方法具有更好的识别准确率。与BP神经网络进行了比较,证明了经验模态分解和最小二乘支持向量机应用于单相自适应重合闸中具有更快的识别速度。
For distinguishing the transient faults and permanent faults on EHV transmission lines,voltage transient signals are decomposed into a number of intrinsic mode functions by using empirical mode decomposition,then the normalized first four IMFs which represent the high frequency（HF（parts of signals as the characteristic values of LS-SVM are used in this paper.A lot of simulation results of EMTP and Matlab calculation show that the method can distinguish the transient faults and permanent faults accurately and shows that the methods has better recognition accuracy.On the basis of the simulation results the method is compared with BP neural network and faster recognition of applying EMD and LS-SVM to single phase adaptive auto-reclosure is proved.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2010年第12期35-39,共5页
Power System Protection and Control
