The High Voltage Transmission Line Fault Location Based On The Hilbert-huang Transform

With the increasing power and voltage grade of transmission line transmission, powersystem security and stability problem become increasingly prominent. When line fault occurs,to find the point of failure and processing it fast, to maintain normal operation of powersystem, has very vital significance.Transmission line failure occurs to produce the travelling wave signals to both ends ofthe line. Along the transmission line, the waves transmit to the transformer substation, whichrecords the waveform characteristics of fault traveling wave. After analyzed, it can carry outwave fault location. The wavelet analysis method is widely applied to the transmission linefault location, but it has some problems. Such as when using wavelet analysis method inpractical engineering, it is need to select wavelet base in advance, which is not according tothe characteristics of the signal itself. As a consequence, it has a great influence on theaccuracy of fault location. Theoretically, wavelet transform method can deal with nonlinearnon-stationary signal. But in the practical engineering, wavelet transform method can onlydeal with linear signal. The scope of its application has also been a certain limit. And also it’sconstrained by the Heisenberg’s uncertainty principle, which has a certain influence on theresults of fault location. While the Hilbert Huang transform fault location does not exist theseproblems.To improve the accuracy of fault location, enhance the adaptability, improve theresolution of low frequency signal, this paper proposes the methods of using theHilbert-Huang transform for fault location. This method bases on the change of signalinstantaneous frequency to search mutation points of signal, according to the signal varies,which has the adaptability, and don’t need select wavelet base in advance. And it is not subjectto the limit of the Heisenberg uncertainty principle, applicable to stationary signals, nonlinearsignals and nonstationary signals, etc, which has a wide applicability. Carry on the EmpiricalMode Decomposition (EMD) for the traveling wave signal fault point firstly, and then apply Hilbert-Huang transform to IMF1to get the signal instantaneous frequency graph. Afterwardsfind out the time of frequency mutation point to identify the wave front of the traveling wave.On the basis of introducing the basic concept and principle of the Hilbert-Huang transform,the simulation example and wave parameters are analyzed. Then find the results of distancemeasuring, carry on error analysis, and compare with wavelet transform.Experiments show that this method for different fault point distance, different fault types,different length of transmission line are efficient and have high precision. It has smallaffection on transition resistance, and has higher accuracy than the wavelet method for faultlocation. It also has strong adaptability.