The Prediction of Gas-in-oil in a Transformer Based on BP Neural Network Optimized by Genetic Algorithm

It will avoid the shortage of maintenance or excessive maintenance that the operational status and the latent faults of a power transformer are effectively predicted. Because of being sensitive to the initial value and easily falling into local minimum, the values obtained from the prediction by BP neural network is not accurate enough. In this paper, BP neural network optimized by genetic algorithm (GA) is used to predict and analyze the gas-in-oil of a transformer. The result shows that the proposed approach can effectively improve the prediction accuracy of BP neural network.