Research On Fault Location Method Of Distribution Network Based On WOA-WPA Improved BP Neural Network

The distribution network at the end of the power system is one of the important links for the distribution of electric energy.In recent years,with the continuous increase of the load connected to the distribution network,the structure of overhead transmission lines becomes more and more complex,and the requirements for the reliability of distribution network are also constantly improved.The distribution network with low current grounding system can run for 1～2 hours with fault after single-phase grounding fault occurs,but if the fault point of fault line can not be determined accurately in time,it is likely to cause interphase short circuit,which brings great hidden danger to the system operation.Fast and accurate fault location of distribution network is very important for the safe and stable operation of the whole power system.This article first for overhead transmission line fault occurs after the current,voltage characteristics is analyzed,based on biot savart law,put forward to detect overhead transmission line fault characteristics of magnetic field changes as the auxiliary method to improve the accuracy of traditional fault location,overhead transmission line fault is obtained by simulation after the current,voltage,and the variation characteristics of the magnetic field.Secondly,the simulation of current,voltage and the change of magnetic field by HHT,effectively extract as overhead transmission lines of the characteristics of the characteristic of multiple samples,application of the BP neural network,the implementation of overhead transmission line fault point location,the simulation found that the more complicated structure of the BP neural network to overhead transmission line fault location has a certain advantage,But there are still some drawbacks.Although BP neural network can meet the requirements of fault location of overhead transmission lines after certain learning and training,its convergence speed is slow in calculation and it may fall into the situation of local optimal solution.In view of the shortcomings of BP neural network,a combination of WPA and WOA algorithm was proposed to optimize BP neural network.The initial weight and threshold of BP neural network were replaced by optimal processing of the characteristic quantity of objective function through WPA-WOA,which improved the operation speed and accuracy of BP neural network.The operation process of WOA-WPA-BP neural network is designed,and a new method of WOA-WPA-BP neural network applied to the fault location of overhead transmission lines is proposed.The feasibility and accuracy of this method for fault location of overhead transmission lines are verified by MATLAB/Simulink simulation.