Study On Transformer Fault Diagnosis And Location

In recent years,with the development of power industry,the installed capacity of power system is increasing,the transformer failure has occurred,at this time,detection and treatment of transformer failure timely is particularly important.In the case of small sample size,support vector machine(SVM)can still solve the nonlinear,high-dimensional and other issues,according to this advantage,this paper established the model of transformer fault diagnosis and fault location model based on support vector machine.The results of fault diagnosis is mainly determined by the parameters of support vector machine,therefore,in this paper,it is proposed that the cuckoo algorithm(CS)to optimize the parameters of support vector machine.The simulation results shows that,compared with other algorithms,this algorithm can improve the accuracy of transformer fault diagnosis and fault location.The main research contents of this paper are as follows:Firstly,in the transformer fault diagnosis model,the cuckoo algorithm is used to optimize the penalty parameter C and the kernel parameter g of support vector machine,in order to improve the optimization ability of the cuckoo algorithm,a new inertia weight co was proposed.We used mixed cuckoo search algorithm and steepest descent to get the new algorithm(SDWCS).This algorithm is used to optimize the SVM parameters,which overcome the defect that the basic SVM model is easy to fall into local optimum.Secondly,in the fault diagnosis model of transformer,a classification model based on support vector machine is established.The support vector machine parameters are optimized by SDWCS algorithm.Support vector machine is trained on MATLAB software platform by using LibSVM toolbox.Through simulation,the fault diagnosis model is compared with Cuckoo Search(CS),Particle Swarm Optimization(PSO),Genetic Algorithm(GA)and Grid Search(GS).Finally,in the transformer fault location model,the transformer oil chromatogram and characteristics of electrical experiments are combined and sent to support vector machine.We construct the multi-level classification model of support vector machine based on binary tree.SDWCS algorithm is used to optimize the SVM parameters,by which the transformer faulat is located step by step.Practice has proved that it can determine the fault location of transformers effectively and increase the accuracy of fault location of transformers,in order to handle the fault frequently.