Transformer Fault Diagnosis Method Based on Improved Cuckoo Search Optimization Neural Network

Abstract

The transformer fault diagnosis method based on the improved cuckoo search optimization neural network disclosed by the present invention first collects the concentration of DGA characteristic gas, and after normalization processing; determines the hidden layer neuron number, training function and The transfer function from the input layer to the output layer is used to establish a fault diagnosis model based on BP neural network; the improved cuckoo search algorithm is used to optimize the parameters of BP neural network to obtain the best weight threshold parameters, and the optimized BP neural network model is obtained. Utilize the training sample to train the optimized BP neural network model, and obtain the improved cuckoo search neural network diagnostic model; adopt the improved cuckoo search neural network diagnostic model to predict the test sample, and its output is the diagnosis result of the transformer failure; the present invention It solves the problems of over-fitting and slow convergence speed of the existing BP neural network, poor solution quality and low diagnostic accuracy in the CS algorithm.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis methods for oil-immersed transformers, in particular to a transformer fault diagnosis method based on an improved cuckoo search optimization neural network. Background technique [0002] The transformer is the core equipment in the power system. According to statistics, the annual failure rate of oil-immersed transformers is 0.00625. Therefore, effective diagnosis of latent faults in transformers is of great significance to the safe and stable operation of power systems; in addition, overheating and discharge faults of oil-immersed transformers are always related to oil Dissolved gas analysis (DGA) technology developed based on this has become an important means of diagnosing transformer faults. [0003] However, in engineering practice, the three-ratio method based on DGA technology has some defects, such as incomplete coding and too absolute coding boundaries. It is an effective way to...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a transformer fault diagnosis method based on the improved cuckoo search optimization neural network, which solves the problems of over-fitting and slow convergence speed of the existing BP neural network, poor solution quality and low diagnostic accuracy in the CS algorithm

Images