Research On The Location Method Of Single Phase Grounding Fault For Distribution Line

Distribution network contact with all kinds of load directly, so it can reflectdemands of electricity for various load quickly and sensitively. With the socialdevelopment and people life improved, in power system the importance of distributionnetwork become more and more higher. Small current earth fault (single-grounding fault)is the highest incidence of failure types in distribution network, So there is an urgentneed to find a fast and accurate fault location method for single phase grounding indistribution network.Compared with the transmission lines, there are more branches in distribution lines,the fault grounding resistance is large and grounding current is small, many positioningmethods were failure. Characteristics for single-grounding fault in distribution network,made a wave c-type traveling wave with artificial neural network fault location method.It integrates the C-type of traveling wave location method with Probabilistic NeuralNetwork. The aim is to combine different location methods to improve the accuracy offault location. In this method, the first step is determining the fault distance by C-type oftraveling wave location method, then because branches which have the same distance tothe busbar may be more than one, so using Probabilistic Neural Network’s patternrecognition feature to further determine the fault branch, which can achieve precise faultlocation. This paper choose the electric wave signal as the analysis signal. In order toobtain more comprehensive results, this paper studies the dead zone and positioningrange of C-type of traveling wave location method. In order to make the simulationwaveform is more close to reality, this paper artificially add "white noise" to simulatewaveform signal to replace the real waveform signal and choose probabilistic neuralnetwork which have more accuracy and stability compared with other neural networks.Build the simulation of the model for distribution network and write the computersimulation programs, the results showed that this method can theoretically achieveaccurate the fault location.