With the development of social economy,the consumption of social power continues to increase,the scale of the distribution network continues to expand,and the requirements for the reliability of power supply are increased.However,the practical operation environment of the distribution network is poor and it is difficult to maintain,besides,various faults are inevitable.Especially arc ground fault,when it occurs,the excessively high temperature of arc will cause insulation damage.Then,the overvoltage of arc will further aggravate the insulation breakdown.If not deal with it in time,a single-phase grounding fault can become a two-phase short circuit easily or even a more serious accident,which seriously threatens the safety and reliability of the power supply of the distribution network.In addition,most of the existing arc fault models ignore the influence of arc length changes.The simulation results are quite different from the practical arc faults in the distribution network.Therefore,it is particularly important to study a more accurate arc fault model,and based on this,to realize quick and effective identification of the arc fault and location of the fault section.(1)Considering the dynamic change of arc length,this thesis proposes an arc fault modeling method for distribution network.First,the arc is discretized into multiple cylindrical small current elements,and the arc motion model is established by analyzing the force of each current element.The change trend of the current element after a fault is simulated,to calculate the arc length at different times.Second,PSCAD and MATLAB is used interactively to build an arc model with dynamically changing arc length.The model is used to simulate the arc voltage,current and resistance waveforms under different fault conditions.Finally,the accuracy of the arc model is verified by comparison with the classic Cassie model,Mayr model and improved cybernetic model.(2)About different types of single-phase grounding faults,this thesis proposes an arc grounding fault identification method based on weighted Euclidean distance,through time-domain analysis of steady-state zero-sequence current.Theory is combined with simulation,this thesis analyzes the influence of neutral grounding method,fault initial phase angle,and fault location on the steady-state zero-sequence current of different types of faults,and summarizes the time-domain waveforms of the steady-state zero-sequence current under different types of faults.Accordingly,the weighted Euclidean distance is used to calculate the similarity between the steady-state current and its fundamental frequency sine wave under different types of faults,to realize the effective identification of arc faults.A large number of simulation experiments have been carried out on the neutral point ungrounded system and the resonant grounded system.The results show that the methods proposed in this thesis can accurately identify the type of fault,and this method is not affected by the initial angle of fault and the location of fault.(3)Based on the accurate identification of arc faults,an arc fault section location method is proposed based on the energy spectrum of characteristic frequency bands,through frequency domain analysis of transient zero sequence current.The S transform is used to analyze the energy spectrum of the transient zero-sequence current of the arc fault,and the characteristic frequency band with the largest difference between the fault section and the non-fault section is determined.Therefore,the fault section is located according to the energy value of the energy spectrum of the characteristic frequency band.The simulation verifies the accuracy of the method under different grounding methods,different initial phase angles of faults and different fault locations.The thesis contains 67 figures,6 tables and 88 references. |