Transformer Thermal Capacity Estimation and Prediction Using Dynamic Rating Monitoring

Abstract

Dynamic rating monitoring is a solution to face challenges in both the operation and the planning of power systems. With thermal monitoring, dynamic overloads of transformers are allowed; however, the thermal model used to predict hot-spot must be capable of ensuring the assessment of insulation life expectancy. Hence, this paper presents an online algorithm to estimate and predict transformers' rating in order to assess the overload capability during short and long times. The overload factors are computed using estimations of top-oil temperature. The main characteristic of this proposal is that the environmental cooling conditions are considered by the fitting of the parameters of a simple equivalent thermal circuit using an extended Kalman filter during the transformer operation. To validate the implementation of the algorithm, the information provided by typical transformer rating monitoring systems is used. Thus, the performance and effectiveness of the algorithm are verified using field measurement data during the normal operation of two power transformers.