Dielectric Loss Factor Forecasting Based on Artificial Neural Network

Now the online monitor and diagnose of capacitive equipments still remains in a simple data processing level. The level of application of diagnosis and monitoring will be improved if advanced mathematical tools for analysis used in it. The paper introduces a new methods of using the BP neural network to predict the dielectric loss angle. Selecting inputs, outputs and parameters of the BP neural network makes it possible to prove the data in experiments. The results show that the method is capable of more accurate to prediction of the dielectric loss angle, and it has a guiding significance to prevent the insulation fault of High-voltage electrical equipment.