Research On Fault Diagnosis Method Of 500kV Substation

With the development of China's power industry,the structure of power system is more and more complex.The lines are closely connected by substations,forming a network structure.In Northeast China,500kV system is the main grid,and 500kV substation is the key node of the main grid.Its line structure is complex and its protection configuration is many.When an accident occurs,a large number of fault information will be generated by the protection and circuit breaker.The monitoring system will transmit these information to the monitoring interface of the substation The operators need to find out the fault point,analyze the fault type and the fault reason timely and accurately according to the fault information.This method improves the accuracy of fault diagnosis based on protection and breaker informationIn order to improve the reliability of fault diagnosis results,this paper takes the 500kV substation of Beining as an example to analyze the fault diagnosis from two aspects of switching value and electrical value.Firstly,aiming at the complex structure and dual protection configuration of 500kV substation lines,when the fault occurs,the amount of fault information is large and the accuracy of information is difficult to distinguish,the time sequence relationship of the substation is introduced into the fault diagnosis analysis,and the time sequence Bayesian fault diagnosis model is established.The received operation information of protection and circuit breaker is verified by time sequence constraints,and the initial probability of each node in the model is determined by combining information entropy theory,then the minimum fault diagnosis result set is calculated by fuzzy reasoning method.This method improves the accuracy of fault diagnosis based on fault information,but it requires a high completeness of switch actionSecondly,in addition to the change of the switch state of the protection and the circuit breaker,the electrical quantities such as the current and voltage at the fault will also have a sudden change.Using the wavelet analysis method to extract the eigenvalues of the electrical quantities to describe the fault degree of the suspicious components,by comparing the eigenvalues of the suspicious components,the fault components can be determined.This method can evaluate and analyze the actual operation of the circuit breaker,but its diagnosis process is too single to distinguish the fault type.On this basis,the information fusion technology is introduced to establish the DS(Dempster Shafer)evidence theory information fusion diagnosis model.The diagnosis results obtained by the two diagnosis methods are taken as the input of evidence body,and the final diagnosis results are obtained through DS fusion decision diagnosis.Compared with the single diagnosis method,the information fusion diagnosis makes the fusion decision of electric quantity and switch quantity,which improves the accuracy of the diagnosis results.The simulation results show that the method is more accurate and effective.