Research On Fault Diagnosis Of Transformer And Cable Based On Live Detection Technology

With the continuous development of the power industry,it is of great significance to ensure the safe and stable operation of power equipment.The fault diagnosis of transformer and cable is an important part of power guarantee.The traditional live detection means have too much information,and there is a certain probability of failure,so it is difficult to carry out comprehensive analysis of transformer and cable.In view of the above situation,this paper constructs a rough set fault diagnosis model to mine the potential laws of things,and introduces the Bayesian network with probabilistic reasoning knowledge to build a joint diagnosis model of transformer and cable fault based on rough set theory and Bayesian network.At the same time,in view of the cognitive uncertainty of data,the regional model and reliability analysis are introduced.In this paper,the principle and application of live detection of transformer and cable are summarized,and the faults of transformer such as discharge,overheat,damp,aging,deterioration and winding deformation are analyzed.The fault condition attributes and decision attributes based on live line detection technology are constructed.Secondly,the rough set theory model is studied,and the reduction and classification of rough set information table are deduced.Finally,a case is given to verify the effectiveness of the proposed method Secondly,the Bayesian network model is studied,and its simplicity,semi simplicity,enhanced simplicity and tree enhanced simplicity are discussed,Then a fault diagnosis model based on rough set and Bayesian theory is constructed,and the fault attribute set and fault type set of transformer and cable based on live detection technology are built,and the fault connection relationship between transformer and cable is calculated,and the Bayesian network is drawn Structure diagram,through the case The accuracy of the model is verified,and the diagnosis results of traditional oil chromatography analysis,rough set theory,Bayesian network and this algorithm are compared by using historical faults,and the results show that the algorithm has higher accuracy;finally,combinedwith the actual application of the field case,combined with the live detection feature analysis technology and the model in this paper,the analysis results show that based on rough set and Bayesian network,the algorithm has higher accuracy The equipment fault diagnosis model of yese theory can indeed assist the field fault diagnosis,but when there is cognitive uncertainty in the data,it will cause misjudgment to the results.Finally,through the introduction of triangular fuzzy number and reliability analysis,the reliability analysis of the judged results is carried out to reduce the impact of misjudgment.