Online Diagnosis And Early Warning For Failure Of Traction Substation Secondary Devices

The rapid development of high-speed railway has put forward very high requirements for the safe and reliable operation of traction power system,among which the secondary devices of traction substation plays a very important role.In order to reduce the negative effects of the secondary devices on the traction power system,it is necessary to study the online diagnosis and early warning for failure of secondary devices in the traction substation.To solve the above problem,this paper makes a comprehensive study which mainly contain the type of failure analysis for the traction substation secondary devices,online diagnosis technology,and fault warning technology.The content is as follows:First of all,the dominant fault of traction substation relay protecti on equipment and hidden fault characteristics are introduced,the transformer measurement value and the different position of transformer measurement based on the relationship between the two methods to determine the possible failure of the equipment,and put forward based on the analysis results of the relay protection and protective device for real-time monitoring,and set the corresponding the threshold scheme,provide data for fault diagnosis and early warning for traction substation equipment.Then,based on the introduction of probabilistic neural network modeling theory,classification of the running state of the transformer,the transformer running state and heating characteristics to the diagnosis of transformer online modeling and analysis.Analysis of BP neural network and probabilistic neural network modeling was used to compare,prove the probability of error of neural network model for transformer on-line fault diagnosis based on the results of smaller,more effective to identify the abnormal operation state of transformer.At the same time,according to the characteristics of the measuring element calculation,analysis of characteristics of distance protection relay protection device,proposes a method for monitoring the impedance measurement,points out that when the adjacent relay measurement difference exceeds the preset threshold,that can run the relay protection device is abnormal,on-line diagnosis device of relay protection.Finally,based on the two traction substation equipment status a nd equipment state monitoring content and technology in detail,and puts forward the basic framework of fault prediction of traction substation equipment two,introduces the data acquisition and processing module,equipment state estimation module and main tenance decision optimization module,and put forward traction substation equipment fault early warning algorithm two times,including two times based on single equipment fault diagnosis based on fuzzy algorithm and fault warning system equipment two Monte Carlo method,and the realization of a detailed analysis of the algorithm.