Power transformer temperature–moisture dynamics modeling: an experimental validation

Abstract

Moisture and temperature are the main factors that affect the life expectancy of liquid-immersed power transformers. These factors speed up the aging process of cellulose insulation, worsening its physio-chemical properties and increasing the risk of faults in the transformer. Recently, a multi-physical modeling approach was reported aimed at estimating the temperature and moisture dynamics of transformer insulation for different loading conditions. The model objective was to evaluate how the moisture dynamics between oil and paper insulation might affect the loading capability of the transformer. Although the model is based on the representation and analysis of concrete regions of the oil–paper insulation, it could also be useful for the estimation of the moisture content of the solid insulation in moisture monitoring systems; however, the modeling had not been experimentally validated due to the difficulty of measuring the moisture in cellulose insulation during transformer operation. In this work, an experiment was designed and developed to carry out the experimental validation of the model and to evaluate the possibility of applying it for moisture monitoring purposes. Moisture dynamics experiments were carried out in an experimental test plant; the evolution of the moisture content throughout the experiment was assessed by direct measure and with dielectric response measurements. The experimental conditions were simulated with the model, and the estimated and measured moisture contents were compared. A detailed analysis was carried out to evaluate the precision of the model and to identify possible sources of error and improvements that may be implemented to improve its accuracy.