| Aiming at the problem that the low accuracy and multi solutions of the power grid fault location caused by the regardless of alarm timing information,and the disadvantages that fault location method of single information source is often inaccurate and fault tolerance is not high,this paper prosed a method taht combines the alarm timing matrix and multi-source information,which is used to solve the fault element and fault time.Firstly,failure elements’ state sequence matrix,alarm time sequence matrix,circuit breaker time sequence matrix,main protection time sequence matrix,near backup protection timing matrix and remote backup protection timing matrix are defined,based on alarm timing information.Also,it defines the method of establishing the elements in the matrix.At the same time,in order to solve the expected value,the solving method of the circuit breaker expected timing matrix,the main protection expectation matrix,the near backup protection expectation matrix and the remote backup protection expectation matrix are defined.Based on the switch and protection information of Supervisory Control And Data Acquisition(SCADA)and wide-area measurement system(WAMS),the timing objective function of fault location is constructed.By using element of the timing matrix of the component state as the state to be solved,a method based immune genetic algorithm is used to solve the timing objective function.It solves the problem that the dynamic alarm information is ignored,and improves the positioning accuracy.Secondly,using SCADA system,fault recorder information system,relay protection system and multi source information of WAMS system to locate the fault equipment of the line.The timing matrix method is proposed by using SCADA switch information,relaying protection information and WAMS switching information;The protection value method is put forward based on the information of relaying protection system;The protection path method is proposed based on the circuit breaker information of SCADA system and protection information of relaying protection system;Recorder method is put forward based on the voltage and current information of fault recorder;The WAMS method is proposed by using the voltage and current information of WAMS system.This paper studied the time alignment method of SCADA,WAMS,relaying protection information system and fault recorder system,solved the problem of time inconsistency,and increased data reliability and accuracy.Finally,the improved D-S evidence theory is used to fuse the multi-source information,which is based on the timing matrix method,the protection value method,the protection path method,the wave recorder method and the WAMS method.The method discount evidence based on determination of importance weights based on Fuzzy Analytic Hierarchy Process,solved the problem that the importance weight can not be determined theoretically.Using possibilistic C-means clustering algorithm to the fusion and classification,the fault and nonfault elements contained in two classes of fault and non fault components are obtained.The accuracy of the results is greatly improved.In order to verify the correctness of the theory,we select an example of a known fault in the literature for verification.The results show that the theory is correct.At the same time,the theory is applied in the oilfield power network.The results show that the fault location method based on the alarm timing and multi-source information can completely identify the fault components of the system.Recognition results are reliable. |
