Research On Distribution Network Fault Location Method Based On Improved CEEMDAN Combined With Frequency Weighted Energy Operator

At present,our country’s power distribution network mainly adopts low-current grounding operation mode.When a single-phase grounding fault occurs,the fault characteristic signal is weak and difficult to detect If the fault location is not determined in time,it may further develop into a phase-to-phase short-circuit fault,which forces the line to trip and cause Large-scale blackouts.Therefore,the precise location of faults is of far-reaching significance for promoting the safe and stable operation of the power system,improving the reliability of power supply and the automation level of the distribution network.Based on this,this paper proposes a fault location method for distribution network based on improved CEEMDAN and frequency-weighted energy operator.A new phase mode transformation matrix is constructed and used for fault analysis,and a single modulus is realized to represent all fault types.When a single-phase grounding fault occurs in the distribution network,the matrix is used to analyze the modulus of the fault transient traveling wave,and the influence of the transition resistance,the initial phase angle of the fault and other fault parameters on the traveling wave characteristics is analyzed from the time domain;The element equivalent model lists the differential equations.From the frequency domain analysis,it can be seen that the line parameters and the traveling wave propagation parameters are functions of frequency after the distribution network fails,and the line mode component is more affected by the dispersion effect than the zero mode component.This provides a theoretical basis for the determination of subsequent fault location methods.A traveling wave fault location method for distribution network based on improved CEEMDAN and frequency-weighted energy operator is proposed.First,the improved CEEMDAN is used to decompose the decoupled line mode components,and the weak fault signal is extracted under the background of strong noise.The fault signal is decomposed into several IMF components,and the correlation-permutation entropy function is constructed.The IMF component is optimized to include the fault sensitive component;then the frequency-weighted energy operator is used to calculate the instantaneous energy of the optimized IMF component,and the time when the incident wave head reaches the two measurement ends is determined according to the first peak point in the time-energy spectrum;A two-terminal ranging method that has nothing to do with the wave speed and eliminates the asynchronous time difference is used to calculate the distance to the fault.A 35kV distribution network simulation system was built through PSCAD/EMTDC,obtain its simulation data,and then write the proposed algorithm in MATLAB to simulate the obtained data.Set up single-phase ground faults under different fault conditions,and conduct simulation verification;at the same time,set phase-to-phase short circuits at different fault locations,and conduct simulation verification on fault location methods.The simulation results show that the fault location method has a small fault location error when single-phase ground fault occurs,and it can still locate accurately in the case of phase-to-phase short circuit.So as to verify the effectiveness and adaptability of the fault location method.