Research On Transformer Fault Diagnosis Based On APSO-BP Neural Network

In China's power system,the power transformer is the most important core component.It plays a very important role in the process of power grid transmission and transformation.In order to achieve safe,stable,and economical operation of the power grid,fault diagnosis It is particularly important,so it is of great research value to use any method to diagnose the transformer in advance.Transformer internal faults are generally analyzed by dissolved gas in oil.This paper chooses BP neural network as the basic algorithm for transformer fault diagnosis.BP neural network has many advantages such as parallel distributed computing,self-adaptation,memory,and clustering.It can accurately express the mapping between dissolved gas in transformer oil and internal faults in transformers.relationship.However,the BP algorithm has the disadvantages of slow convergence speed and easy to fall into local minimum points.Therefore,the particle swarm optimization algorithm is selected to optimize the BP neural network.The particle swarm algorithm has the ability of global optimization,which can effectively improve the convergence speed of the BP neural network and improve the accuracy of the fault.Rate;but the standard particle swarm algorithm is easy to fall into the "premature"phenomenon during the optimization process,and into the local optimum.The APSO-BP algorithm is introduced into the particle adaptive function to adjust the particle calculation,which can solve the local appearance of the standard particle swarm algorithm.Optimal problem.In this paper,a BP neural network transformer fault diagnosis system with 5-10-5 as the neuron structure is established according to the characteristics of the fault gas inside the transformer fault tank and the fault type.The PSO-BP neural network transformer is established using the particle swarm algorithm to optimize the BP neural network.The fault diagnosis system,and the APSO-BP neural network transformer fault diagnosis system is established by optimizing the standard particle swarm algorithm.After training and results,the unclassified BP neural network fault diagnosis model for transformer fault classification is 82.5%,the PSO-BP neural network fault diagnosis model is 90%,and the APSO-BP neural network fault diagnosis model is 95%.By comparing the accuracy of the three algorithms,it is concluded that the APSO-BP transformer fault diagnosis system can more accurately diagnose fault data.