Cooling performance of different dielectric fluids containing nanoparticles in a transformer winding

Abstract

This work presents a study where the thermal performance of different nanofluids is tested in a transformer winding model. A 2D axisymmetric non-isothermal CFD model has been used to compare temperature and velocity distribution of the proposed fluids. The fluids tested consists of eight different mixtures manufactured from two different nanoparticles and two different concentrations, using a mineral oil and a natural ester as base liquids. The comparison has been carried out between each nanofluid and their base fluid to observe the thermal impact of nanoparticles. The temperature dependent properties have been determined to include them in the simulation model. Two different inlet velocities were selected as inlet conditions, representing Oil Natural and Oil Directed Cooling. The analysis has been carried out using ANSYS Fluent ® in a two pass winding model. The average and maximum disc temperatures and mass flow distributions are obtained for all the cases. The results obtained show that the temperature in the seven discs of the considered winding increases when the nanoparticles are added to the natural ester, between 3 and 7%. On the other hand, in the case of mineral oil, the addition of nanoparticles reduces the temperature in the discs between 0.3 and 3%.