Based On Morphological Analysis Of The Small Current System Ground Fault Identification Technology And Engineering Applications

The non-effective grounding system is used in the distribution net of 6-66KV. It has the virtue of proving power supply's reliability. But ground fault identification and line selection are not solved when single-phase earthing appears. This paper studies single-phase earthing fault identification and its engineering application in the non-effectively grounding system.This paper analyzes the phenomenon and mechanism of the single-phase-to-ground faults and the resulting fault such as ferromagnetic resonance, high voltage fuse of PT blowing, and over voltage of asymmetric break. Combined with the 10kV fault selection simulation test system and different types of simulation in single-phase-to-ground faults, this paper has collected numerous fault waves in the fault identification.This paper proposed a comprehensive line selection program named "To select the ground fault identification algorithm based on the fault process analysis", which incorporates a variety of line selection criterion for line selection. Starting from the the state characteristics of the faults, this paper have a detailed analysis the transition changes (i.e. changes in the characteristics from transient to steady state), successively after the single-phase-to-ground faults occur. With the extraction of morphological characteristics and harmonic feature based on the with mathematical morphology analysis, this paper proposed the fault identification criterion based on the process analysis of the failure.This paper indicates that in the non-effective grounding system, the overall process of development of the ground fault can be seen as constituted by a series of sub-processes, which are transient state, steady state and temporary state. It is also pointed out that it is possible to achieve the purpose of fault identification by distinguishing the time order of different sub-processes and the own state of those processes.Utilization the technology of wavelet analysis and mathematical morphology, this paper constructs the nonlinear form of wavelet and proposes a new method of fault identification. This method introduces the morphological into the transferring of the zero-signal after the failure, to extract fault quantify factor characterized by the harmonic factor and shape factor, and the diagnostic model is formulated. This paper established the feature set during power failures of both neutral point grounding system and neutral grounding via arc-suppression coil, such as single-phase-to-ground fault, series resonance, ferromagnetic resonance and other characteristics of the fault process and the processing of feature extraction is also describes in this paper . This paper describes the structure and on-site application of ZXJ-11 integrated fault diagnosis and wave recorded controlling device developed by our project team. And some on-site typical recorded data are also analyzed in this paper, which shows the correction of the fault characteristics identification proposed in this paper.