Electronic International Standard Serial Number (EISSN)
In the last years many works have been published dealing with the importance of modeling the behavior of moisture inside the transformer. The most precise models are those that analyze the moisture transport processes between solid and liquid insulation by means of Fick's diffusion theory. The use of these kinds of models requires to know the moisture diffusion coefficients of the analyzed materials. In this paper the experimental methodology used and the results obtained in the determination of the moisture diffusion coefficient of oil-impregnated transformer insulating paper are presented. Experimental drying curves of paper samples were obtained, and then the diffusion coefficient were found applying an optimization process based on the finite elements method and genetic algorithms. As a result, four new expressions for the moisture diffusion coefficient of impregnated insulating paper are proposed, that take into account the dependence of the diffusion coefficient on variables such as temperature, local moisture concentration, insulation thickness and ageing degree of the insulation paper. Finally the simulated drying curves obtained by using the proposed coefficients, and using others coefficients from the technical literature, were compared with drying curves obtained experimentally.
power transformers; cellulose insulation; moisture diffusion coefficient