Research On Fault Line Selection And Overvoltage Based On Distribution Line Status Online Monitoring System

Chinese low-voltage power distribution network widely adopts small-current grounding mode(neutral point is not grounded or arc-extinguishing coil grounded)to operate.Because of its complex structure and varied operating environment,distribution networks have frequent failures,of which Single-Phase-to-Ground faults occur most frequently.According to the relevant regulations,the system is allowed to continue to operate for 1-2 hours after a Single-Phase-to-Ground fault occurs,but the Single-Phase-to-Ground fault will generate overvoltage.Long-term faulty operation will bring serious threat to the safe operation of the system.Therefore,it is very important to identify faulty lines in a timely and accurate manner.With the development of distribution automation,distribution line status online monitoring systems are increasingly being put into operation to ensure the safety of distribution networks.The system can obtain a large amount of system operation data.These data provide new ideas for fault line selection and can be used to evaluate overvoltages.Firstly,This paper analyzed the fault characteristics of the Single-Phase-to-Ground fault in the small current grounding system in detail,and revealed the shortcomings of using the steady state information after the Single-Phase-to-Ground fault occurs to select the fault line.In view of this,This paper proposed a method for fault line selection using transient energy after a Single-Phase-to-Ground fault.This method is based on the distribution line status online monitoring system using this system to collect the transient zero-sequence current after a Single-Phase-to-Ground fault.This method decomposes the transient zero-sequence current generated after a Single-Phase-to-Ground fault occurs on each line,and then constructs the relative energy factor of each line.By comparing the relative energy factors of each line,the fault line is identified.The correctness of this method was verified by simulation experiments and analysis of the actual data collected by the distribution line status online monitoring system.Secondly,this paper introduced compressed sensing theory into fault line selection and proposed to use Bayesian compressed sensing to replace traditional sampling.In the sparse Bayesian framework,the sampling signal is recovered through reconstruction algorithm precisely.Therefore,this paper used Bayesian compressed sensing to accurately reconstruct the transient zero-sequence current signal,based on this,the relative energy factors of each line are calculated to achieve fault line selection.This paper validated the correctness of the method by using Matlab/Simulink software for modeling and simulation.Finally,due to the fact that the distribution line status online monitoring system can count information such as overvoltage duration,on this basis,this paper proposed a method to evaluate the overvoltage caused by Single-Phase-to-Ground faults.Calculating the probability density curve of the overvoltage duration by non-parametric kernel density estimation method.From this curve,the distribution of overvoltage duration can be reflected.Besides the average value of overvoltage duration can be further calculated to realize overvoltage evaluation.In this paper,the overvoltage was evaluated by using non-parametric kernel density estimation method and the actual data collected from the distribution line status online monitoring system.This result provides a certain theoretical basis for the insulation protection of the distribution network and Power quality assessment.