Study On Self-Synchronization Method Of Fault Data In Smart Substation

Synchronization of sampling data,especially for fault data,plays a decisive role in correct operation of current differential protection in smart substation.Besides,it has an important impact on the accuracy of fault location,fault recording and fault analysis.Conventional methods for synchronization include clock synchronization based on simultaneous sampling by merging units,and interpolation algorithm based on free sampling.The above schemes have disadvantages of the complicated process,high cost,and insufficient stability.In some case,the data may lose its synchronization due to technical defects of the synchronization loop,which has an adverse influence on protection's behavior.Therefore,it is of great practical significance to solve the synchronization problem of fault data for the correct operation of the protection device and the safe operation of the power system.The main work of this paper is as follows.Firstly,this paper introduces the necessity and basic principles of data synchronization.The necessity of exploring self-synchronization methods can be clarified by exploring the advantages and disadvantages of widely used synchronization methods.Secondly,the basic principle of the self-synchronization method of fault data is described.Error analysis and feasibility evaluation are performed on the traditional self-synchronization method based on start instant of the protection.Furthermore,it is clear that the key to the popularization and application of the fault data self-synchronization method is to perform accurate fault moment detection.Thirdly,in this paper,a fault-instant detection method based on curve fitting is implemented.By solving the zero-crossing point of the fitting function of the phase current increment,the fault instant can be extracted.In order to solve the multi-solution problem generated during the detection,an improved method of waveform similarity is proposed.Analysis shows that the detection error will not exceed one sampling interval.Fourthly,a fault instant detection method based on multi-threshold discrimination is proposed.By setting multiple thresholds,point-by-point discrimination and selection of phase current mutation amount are performed to find the fault starting point.The operation is simple and the calculation time required by the algorithm itself is negligible.It can extract fault instant at the moment of fault detection element startup,which is helpful for real-time query of fault information,quick processing of fault data and rapid action of relay protection.Finally,through the PSCAD simulation and actual wave example,the synchronization error of the proposed self-synchronization method is analyzed.Furthermore,the proposed self-synchronization method is applied to the main protection of transformers,bus bars,outlets and the backup protection of the smart station.The performance is analyzed compared with the traditional self-synchronization method.The test results verify the feasibility of the proposed self-synchronization method.