Research On Fault Line Selection For Single Phase Grounded Fault In Small Current Grounding System Bsaed On The Multivariate Information Fusion

In the medium and lower voltage distribution network of our country,the neutral ungrounded mode or the arc-suppression coil grounded mode is widely used,which is known as the small current grounding system.When the single-phase ground fault occurs in the small current grounding system,the fault current is very small and the line voltage is still symmetrical,so allow keep power supply 1~2 h,which improves the reliability of power supply.However,due to the small fault current and other factors,the detection accuracy of single-phase ground fault in small current grounding system is low.Therefore,how to quickly and accurately detect single-phase ground fault has become an important issue to improve the reliability of the small current grounding system.Aiming at the problem of single-phase ground fault line selection in small current grounding system,this thesis reviews many domestic and foreign literatures,and summarizes the research results of predecessors.According to the different use of signals,the existing methods of grounding line selection are classified,and then the advantages and disadvantages of various line selection techniques are analyzed and compared.At the same time,the neutral ungrounded system and arc-suppression coil grounded system were modeled and analyzed,the oscillation mechanism of transient current in single – phase ground fault is studied by using phase-mode transformation theory.Finally,we get the fault characteristics of steady-state signal and transient-state signal,and then select the fusion line selection method by combining the steady line selection methods and transient line selection methods.For line selection method based on steady-state information,this thesis selectes fifth harmonic method and active power method,which are not affected by arc-suppression coil.Fast Fourier transform is used to extract the required fault characteristics,and the fault line is identified through through comparing the fifth harmonic and active power amplitude and direction of each line.For line selection method based on transient signal,this thesis selects wavelet packet transform as the fault line selection algorithm based on analysis and study of wavelet transform theory,and gives the line selection process.At the same time,a small current system model,included overhead lines and cable lines,is built by using SIMULINK simulation platform in MATLAB.Under different conditions,the simulation results show that the methods have high accuracy,but there still possible has mistakes at extreme conditions.Therefor,the BP neural network is used to combine the steady-state information and the transient-state information.Through the study of neural network theory,amplitude and direction fault measure of steady signal and transient signal are defined as the input of the neural network,and the result of the selection line as the output of the neural network.Through simulation under different fault conditions,and compared with single line selection method,the results show that the line selection method based on information fusion can greatly improve the line selection success rate of single-phase ground fault with strong anti-interference ability and high reliability,and has important theoretical significance and high practical value for studying small current grounding system...