Font Size: a A A

Research On Single-phase Grounding Fault Location Of Distribution Network With Distributed Power Generation

Posted on:2021-04-28Degree:MasterType:Thesis
Country:ChinaCandidate:Q S QinFull Text:PDF
GTID:2432330602471203Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
As the permeability of DG in distribution network increases gradually,the power flow characteristics and grounding fault characteristics of distribution network are complicated to a certain extent.The topology and power flow distribution of the medium-low-voltage distribution network have been changed,which makes the single-power supply network into dual-power or even multi-power network.Thus,fault line selection and location method based on traditional distribution network cannot be applied to distribution network with DG.In this paper,the fault fault selection and location method of distribution network with distributed generation is studied to reduce the outage cost,and the following work is carried out.Firstly,the current situation of fault line selection and location at home and abroad is summarized,and the steady-state and transient characteristics of resonant grounding system fault are derived.The defects of traditional fault location method are illustrated by a simple example.IIDG topology and grid-connected rules are introduced and its output current characteristics under normal operation and single-phase grounding fault are analyzed.Based on Matlab/Simulink,this paper establishes the equivalent model of distribution network.Simulation results indicate the validity of the theoretical analysis of zero-sequence current.Secondly,a CEEMD-DTW fault line selection method is proposed.The zero-sequence voltage limit of the bus is taken as the starting criterion of the line selection process,and the zero-sequence current of the headend of each feeder is extracted for CEEMD decomposition before and after the fault.The linear correlation coefficient analysis is carried out by taking the highest frequency inherent mode function to determine the suspected fault line.Then,the imf of the suspected fault line is selected for correlation analysis to set the characteristic frequency band and its corresponding characteristic imf components.In the characteristic frequency band,the normalized DTW distance between each feeder characteristic imf component is calculated to obtain the fault line.The line whose DTW distance exceeds the threshold is judged as faulty line.Based on the simulation of different fault factors and DG location and capacity,the correctness of fault line selection method is verified.Finally,for complex multi-feeder distribution network,a fault location method based on outliers weighting and DBSCAN clustering is proposed.This paper fuse a variety of fault feature quantities,and generate sparse high-dimensional fault feature state monitoring matrix.Use MDS-SVD to reduce the high-dimensional fault feature state monitoring matrix to twodimensional space.To further differentiate the node state,this method calculates the outlier factor and multiply it with the dimension reduction matrix,and combine the DBSCAN clustering method to realize visualization and fault section location.The variation law of the weight coefficient of outliers under different fault resistance and initial phase angle is obtained by simulation experiment.The simulation results of different fault conditions show that the proposed fault location method can accurately and quickly judge the single-phase grounding fault section,which has strong sensitivity.
Keywords/Search Tags:distribution network, distributed generation, single-phase ground fault, CEEMD-DTW, outlier factor weighting, MDS-SVD
PDF Full Text Request
Related items