Study And Implementation Of Traveling Wave Based Fault Location Method In Distribution Network

A complete network of production and consumption of electricity, includes energy generation, transmission, distribution and being used also. The distribution network takes the direct link with the user, and assign the duties of power transmission terminal assume. However once a failure occurs, it will cause a short-term or long power outage and lead people’s lives to inconvenience. On the other hand, the branches of distribution network are so many that it is difficult to quickly locate the point of failure, with a waste of time and manpower. So the research of fault location in distribution network is significant.When line faults occur, the voltage transients relate to the fault locations. Fault-originated travelling waves propagate along the distribution paths and are reflected at discontinuity points of impedance. The thesis first proposed the method that uses continuous-wavelet transform (CWT) for the analysis of voltage transient duo to line faults to get the characteristic frequency which would indicate the fault location. On the base of the mentioned method, the thesis look for ways to improve by taking the advantage of the computing efficiency of FFT. Finally, a fault location method based on artificial neural network by the characteristic frequency of the transient voltage is illustrated. After learning the frequency spectrum of voltage transient originated by fault in different location of the distribution power system, the ANN would seek out the relationship between the frequency spectrum and the fault location.Simulations are performed respectively using PSCAD and MATLAB. The result showed that the proposed algorithm has high accuracy and has no relationship with velocity of the travelling wave. This algorithm could also be applied to various fault situations, and promising in the future.