Research On Power System Fault Signal Processing And Recognition Method

When power system occur faults,there will be complex electric signals,including steady state and transient signal,especially,transient signal contains very rich information,how to correctly deal with these signals,which is very meaningful to study the application of the electric power system fault detection,fault identification and diagnosis and relay protection.Here,the problem of fault transient signal processing to solve is how to extract the characteristic information of fault from fault signal.There are two difficulties in the process,firstly,the fault transient signal has the characteristics of small amplitude,low energy,easy to be masked by steady signal and noise,which is not easy to be detected;the second is that the fault transient signal is very rich,complex,often can not be directly characterize fault state or type.Aiming at these problems,the information entropy theory and evidence fusion theory be introduced into the power system fault signal processing and feature information.This paper studies the method of processing and analysis of power system fault signal,and apply to the identification of magnetizing inrush current and fault line selection in small current grounding.This paper introduces same methods of fault signal processing used in power system for extracting feature information,briefly discusses its advantages and disadvantages,and application of information measurement theory framework,analysis the main information index,aiming at the limitation of information measure,and build the model of information fusion using multiple information measure index.Inrush current is the main factor which affect the correct operation of transformer protection.the problem of magnetizing inrush and internal fault current recognition is introduced,briefly.According to the characteristics of inrush current and fault current respectively,combining the information entropy theory and wavelet analysis method,extraction method is proposed based on the time-frequency features of fault information space,and get the wavelet information entropy indexes effectively,finally,this method is applied to identification of magnetizing inrush and internal fault current,and the simulation results are given to verify the validity of the proposed method.Fault identification and small current grounding line selection has been one of the hot issues to research in recent years?Proceeding from theory of information entropy,combined with DS evidence fusion theory,information fusion method based on wavelet information entropy measure is proposed.Process is that the character of the fault signal is extracted by using a variety of different wavelet information entropy firstly,and respectively as independent evidences,and then fuse evidences of feature to obtain characteristic parameters,according to the trusted number decision,reach the correct,reliable purposes of fault identification.And fault simulation experiments verify the reliability and accuracy of the method.This thesis makes a comprehensive and in-depth theoretical study and simulation verification.The research of this paper shows that this new method of power system fault signal analysis and processing has a certain theoretical significance and practical application value.