A new dynamic model for propagation of characteristic gases in transformer oil-cellulose structures due to temperature variations

Dissolved gas analysis (DGA) is one of the best known and most effective ways of monitoring the condition of power transformers. However, since transformers are rarely in thermal equilibrium, migration of characteristic-gases into insulating cellulose of transformers due to temperature variations causes error in dissolved gas analysis. This paper suggests a new diffusion dynamic model to explain the migration of transformer characteristic gases in an oil-cellulose structure. An innovative method has been used to obtain the model parameters from limited experimental data. The method is based on the measurement of diffusion times of characteristic gases inside the oil-impregnated cellulose. Measured data is then processed and identified as a linear system in transfer function form. The diffusion parameters of characteristic gases are obtained analytically from the identified model. Finally, these parameters are used in the suggested diffusion dynamic model, and the model is solved for a temperature step change.