Fault Diagnosis Methods For Power System Using Multi-dimensional Information

Extreme weather and equipment failure has been a threat to the safe and stable operation of the power system,which will lead to a large area of power outages under some serious cases.Therefore,how to accurately and quickly carry out the power grid fault diagnosis in order to provide the operation dispatchers the diagnostic results can help them to timely deal with the fault,shorten the accident processing time,in order to prevent the expanded accident as well as reduce the loss of power.It’s of great significance to ensure the safe and stable operation of the power system.At present,the fault diagnosis method mainly uses the data provided by the supervisory control and data acquisition(SCADA)and the fault information system(FIS).However,these data are not consistent with the time scale,and the flooding of the fault information is caused by the large influx of the data into the dispatching center in complex fault cases.With the application of the wide-area measurement system(WAMS)in the power system,it’s possible to provide its unified time-scale wide-area electrical information for the fault diagnosis.In addition,the multi-dimensional information of different data sources can be integrated by using the WAMS unified time scale data,which can provide richer fault information for fault diagnosis of power system and lay the data foundation for improving the accuracy of power system fault diagnosis method from the point of view of data source.In this context,this paper introduces the overall framework of fault series and integration platform based on multi-dimensional information to apply WAMS data to the power system fault diagnosis.The four key technologies of this platform have been research.The main progresses are as follows:(1)Aiming at the problem of bad data in WAMS data,on the basis of summarizing the common types of WAMS bad data,a fast identification and recovery method of WAMS bad data based on pattern recognition is proposed.The simulation results show that the proposed method can quickly and effectively identify WAMS bad data and repair it correctly.(2)In order to local the fault time and fault area rapidly,a fault time determination and fault location method based on random matrix theory is established.This method realizes the extraction of fault time and the location of fault area by using the single ring theorem in random matrix theory.The simulation results show that the proposed method can accurately and easily determine the fault time and locate the faulty fault area even when the WAMS data contains the measurement noise.(3)In order to improve the existing WAMS fault diagnosis method based on pattern recognition,which is difficult to obtain the detailed fault developing process,a fault diagnosis method based on WAMS time series information is proposed to obtain the whole fault developing process by introducing the timing information into the standard pattern vectors.The simulation results show that the proposed method can accurately diagnose different fault types and obtain the detailed fault developing process.(4)In order to fully exploit the application of multi-dimensional information including SCADA,WAMS and relay protection information system in the fault diagnosis of the power system,a fault diagnosis method based on multi-source information time series matching is proposed.The feasibility and validity of the proposed method are verified by the multi-source information collected from the actual faults.