Research On High Impedance Fault Identification Method Of Distribution Network

Distribution network,as an important part of the connection between the large grid and the user’s load side,is a key link to ensure the quality of power supply and improve the stability of the power system.In the context of new power system construction in our country,how to improve the reliability and security of power supply has become an urgent problem to be solved in modern distribution network.Because the distribution network covers a wide area and operates in a complex environment,it is easy for the distribution network lines to contact with the ground or other media,resulting in high impedance faults.When a high resistance grounding fault occurs,its fault current is small,the electrical quantity changes are not obvious,and it is difficult to be recognized by existing line protection devices.It is easy to cause major safety accidents such as electric shock to the human body,seriously threatening the safe operation of modern distribution networks.Therefore,this article studies the identification method of high resistance grounding faults in distribution networks.In this paper,the influence of different fault conditions on high impedance fault waveform is fully studied through modeling and simulation,and an adaptive signal decomposition method suitable for high impedance fault identification is proposed,which improves the decomposition effect of high impedance fault data,and a high impedance fault identification method based on multi-energy moment feature fusion is designed,which improves the accuracy of high impedance fault identification.The main research content of this paper is as follows:(1)Analysis on the Influence Factors of Waveform Form and Energy Distribution of High Impedance FaultThe influence of different fault conditions on waveform and energy distribution of high impedance fault is analyzed by modeling and simulation.The Emanuel model is selected to simulate the high resistance grounding fault,and the distribution network model with distributed power is built with the neutral point grounded through the arc suppression coil.The zero-sequence current and timefrequency energy distribution of high-resistance ground faults under different initial fault phase Angle,different fault location,different line types and different distributed power output are simulated and analyzed,and the influence of different fault conditions on waveform form and energy distribution is summarized.The simulation results show that various fault conditions have little influence on the energy distribution of high frequency component of high impedance fault,which can be used to design fault characteristics and identification model.(2)Waveform Decomposition Method of High Impedance Fault based on Adaptive Variational Mode DecompositionIn order to accurately decompose and extract the high frequency components of the high impedance fault waveform,an adaptive variational mode decomposition method is designed based on the variational mode decomposition.In view of the defect that the parameters of variational mode decomposition need to be set manually,an adaptive weighted African vulture optimization algorithm was proposed to optimize the parameters of variational mode decomposition and reduce the error caused by the lack of prior knowledge.Considering the high entropy and complexity of waveform information of high impedance fault,a composite evaluation index suitable for high impedance fault field was constructed as a fitness function to improve the completeness and orthogonality of decomposition results.The experimental results show that the proposed decomposition method has the characteristics of fast convergence,high resolution and high degree of orthogonality.(3)High Impedance Fault Identification based on Multiple Energy Moment FusionUsing the decomposed high frequency components,a high impedance fault identification method based on multi-energy moment feature fusion is designed.Firstly,the data of zero sequence current is preprocessed by difference operation,and the waveform is decomposed by adaptive variational mode decomposition to extract the high frequency component of zero sequence current.Then,based on the energy moment characteristics,three fault characteristics,namely,half-interval energy moment ratio feature,high and low frequency energy moment ratio feature and energy moment entropy feature,were constructed.Finally,the constructed fault features are identified as input vectors of median dendritic neural network to improve the accuracy of high impedance fault identification.The experimental results show that the proposed fault identification method has good anti-noise ability and practicability,and can realize the accurate identification of high impedance faults.