Research On Transformer Fault Diagnosis Based On Grey Wolf Algorithm Optimization Probabilistic Neural Network

Transformer is a very important component in power system,which can convert high voltage electric energy into low voltage electric energy and play a key role in power transmission and distribution.The running condition of transformer is directly related to the stability and reliability of the whole power grid,so the establishment and improvement of transformer fault diagnosis system is an indispensable part of the power industry.At present,the research focus of transformer fault diagnosis system is to improve the ability of real-time monitoring and the accuracy of diagnosis.In recent years,with the construction of smart power grid and the development of data acquisition technology,data-driven transformer fault diagnosis method is widely used in practice.This method is based on the historical operation data of the transformer,and the model is built by data mining and machine learning techniques,so as to improve the accuracy and efficiency of fault diagnosis.The significance of transformer fault diagnosis can not be underestimated.It is the key to ensure the safe and reliable operation of transformer equipment and ensure the quality of power supply.In order to improve the fault diagnosis accuracy of oil-immersed power transformer,this paper firstly conducts an in-depth study on the transformer fault types and the mechanism of gas generation,establishes a probabilistic neural network transformer fault model,and optimizes the PNN fault diagnosis model based on grey Wolf algorithm,and establishes a GWO-PNN transformer fault diagnosis model.On this basis,GWO algorithm is further optimized based on CAT chaos and Gaussian variation,and a CGGWO-PNN transformer fault diagnosis model is constructed.The model combines CAT chaos and Gaussian variation optimization,and the new algorithm optimizes the weights and other parameters of the neural network.CAT chaos can also effectively balance the global search and local search capabilities of the Grey Wolf algorithm.The algorithm can avoid falling into the trap of local optimal solution effectively and improve the calculation accuracy and convergence speed of the algorithm.Through the simulation and analysis of three kinds of transformer fault diagnosis models by Matlab,the excellent performance of CGGWO-PNN transformer fault diagnosis method in the aspect of fault accuracy and accuracy is verified,and it is an efficient transformer fault diagnosis method.At the same time,this paper also builds a hardware platform and develops a transformer fault diagnosis system based on Lab VIEW,which more reflects the integrity of the research content of this paper.