Absrtact:Aiming at the problems of slow response speed and low recognition accuracy in fault type identification of grounding lines in power system,a fault type identification method based on signal processing,pattern recognition and data fusion is proposed.Firstly,wavelet transform and empirical mode decomposition are used to extract the time-frequency domain features of fault current and voltage signals.Secondly,support vector machine and probabilistic neural network fault identification models are constructed respectively.Finally,the diagnosis results of the two models are fused by evidence theory,and the final fault type judgment is obtained.The experimental results show that this method can effectively identify typical fault types such as single-phase metallic grounding,single-phase high-resistance grounding and two-phase short circuit,with high identification accuracy and strong reliability,which provides a strong support for ensuring the safe operation of power system.
关键词
电力系统/接地极线路/故障类型识别/信号处理
Key words
power system/grounding line/fault type identification/signal processing
维普
万方数据