Study On Fault Location Method In Non-effectivel Grounded Network And Its Application

Non-effectively earthed system, mainly including neutral point ungrounded and grounded via arc suppression coil, is widely used for MV distribution network in many countries. Single phase-to-earth fault is the most common fault type in distribution network. The distribution network can run for a period after the single phase-to-earth fault (also known as the small current grounding fault) occurring and some instantaneous faults can recover automatically. But the single phase-to-earth fault can produce over-voltage, which will damage the system security and even lead to power interruption. The location of fault point must be determined quickly to ensure the system security and reliability of power supply. For a long time, selection of the fault line and location of the fault point are very difficult in non-effectively earthed system because fault current is very weak and fault arc is often intermittent. Now the technology of fault line selection has almost matured and achieved satisfactorily effect in field applications. But the difficulty of fault location has not yet been resolved very well. Most fault location technology is still in the phase of theoretical research and lack of actual field applications. The location difficulty of single phase-to-earth fault has plagued electric power enterprise for a long time, which restricts the development of distribution automation technology strongly and must be solved as soon as possible.The single phase-to-earth fault location method using signals generated by the fault itself is researched. The main research works and results in this paper are as follows:(1)The present situation of single phase-to-earth fault location technologies and main problems are analyzed. The fault location technologies are still in the phase of theoretical research, most of the existing method is just feasible in theory, but the key technology of field application has not been effectively resolved, failed to achieve practical application level. The main problems faced by fault location method are applicable range not wide enough, difficult to acquire the required signals, difficult to synchronize precisely between FTUs, etc.(2)A novel fault location method based on zero-sequence admittance of section is presented. Zero-sequence admittance of section is defined as the zero-sequence measured admittance difference between the two sectionalized switches of that section. The difference of zero-sequence admittance between the sound section and fault section is analyzed. The fault section criterion based on zero-sequence admittance and process of fault location is presented.(3)Transient characteristic difference between upstream and downstream of fault point is analyzed. The three-phase system is transformed into module system by using Karrenbauer transformation. The circuit model of earth fault is obtained by simplifying the mode-impedance. Research results show that transient process of upstream and downstream of fault point are independent, and the transient zero-mode current characteristics is determined by parameters of the zero-mode network.(4)A novel fault location method using transient line-voltage and transient zero-mode current is proposed. The phase relationship between transient line voltage and transient zero-mode current is analyzed. The method to monitor earth fault and determine fault phase by using line-voltage is proposed. The mean value in transient period of the product of transient line-voltage after Hilbert transformation and transient zero-mode current is defined as the fault direction parameter. The direction parameter calculated by FTU downstream the fault point is greater than zero, but that calculated by FTU upstream the fault point is less than zero. According to this characteristic, master station can determine the fault point location by analyzing direction parameters uploaded by FTUs.(5)The fault location method based on transient zero-mode current waveform similarity is improved. The improved method takes into account the correlation coefficient polarity, determining section that correlation coefficient less than zero is fault section. The method calculating maximum correlation coefficient with waveform panning is adopted to reduce the effect of asynchronous sampling between FTUs. The correlation coefficient result is corrected by using difference of current signal amplitude. Correlation coefficient threshold value is set self-adaptively according to the correlation coefficients of all sections of the fault line. If the section correlation coefficient is greater than threshold value, it is sound section, otherwise it is the fault section.(6)The fault location methods are validated using simulation data and field fault experiment data. Simulation models are built by the ATP-EMTP software, and simulation data are processed and calculated by the Matlab software. The correctness of location methods presented in this paper is validated by the simulation and experiment results.(7)Location system for single phase-to-earth fault is designed to put into operation in the field. The method based on transient zero-mode current waveform similarity is adopted by the location system. The location system is validated by the artificial grounding experiment and put into practical operation in the field. Field experiment and operation results show that the location system can achieve accurate location for single phase-to-earth fault and meets the field operation requirements.The fault location methods proposed in this paper are realized based on the distribution automation (DA) system, with the advantage of free from influences of arc suppression coil, safe, economical and easy to implement. Section zero-sequence admittance method does not need high speed sampling and complicated calculation, with a low requirement for hardware processing ability of FTU. Method using transient line-voltage and transient zero-mode current with a high reliability, can be applied to situation that FTU has the line-voltage transformers. Transient zero-mode current similarity method only needs to compare the zero-mode currents, thus it meets the condition that only the zero-mode current signal can be measured. The validation results by simulation and experiment data show that proposed location methods are feasible and provided with the condition of practical application. Transient zero-mode current similarity method has been put into actual operation and can meet the field requirements for fault location. The application of study achievements in this paper is expected to solve the technical problem of reliable location for single phase-to-earth fault in distribution network.