Research On Single Phase To Ground Fault Line Selection Device For Small Current Grounding System

With the continuous development of the economy,the demand for power supply in our country is increasing.To ensure the reliability of power supply,to minimize the number of faults in distribution network and to select and maintain the line in time after the faults have become the problems facing the power personnel.Distribution network in China has various operation modes,complicated lines and many random factors,which interfere with fault information collection.At present,the accuracy of all kinds of line selection and positioning devices is not high.In this paper,single-phase-to-ground fault line selection method and line selection device for small current grounding system are studied,and a new line selection method is proposed.Finally,the BP neural network is used to fuse line selection,and the realization of hardware device is introduced.Firstly,the transient and steady-state characteristics of neutral-point grounded system and neutral-point grounded system via arc-suppression coil and their respective faults are studied.The changes of zero-sequence signals of the system when faults occur are analyzed.Various traditional line selection methods based on transient and steady-state characteristics are introduced and their applicability is analyzed.And good and bad.Then three traditional routing methods are selected and simulated in Matlab to verify their effectiveness and analyze their limitations.Secondly,a fault line selection method based on empirical wavelet transform for single-phase-to-ground fault in small-current grounded system is proposed to solve the problem of difficult or even mistaken fault line selection due to the lack of obvious fault characteristics in single-phase-to-ground fault line selection of small-current grounded system.The effectiveness and superiority of the proposed method are verified in the Matlab simulation software.Finally,the process of realizing the method of line selection based on BP neural network is expounded,and the network model of line selection based on BP neural network is built.The transient and steady eigenvalues are extracted by four methods,namely zero-sequence power,fifth harmonic,wavelet packet modulus maxima and empirical wavelet,as the input of BP neural network,and the test samples are used.Simulation results show that the proposed method can realize fault line selection of small current grounding system and improve the accuracy of line selection.Secondly,the hardware and software of the fault line selection device of the small current grounding system is designed.