N Najafi
M Allahbakhshi
M Bagheri

Optimal performance of a power transformer during its normal life depends on the quality of its insulation system. As the load increases, there is a risk of fault due to the increase in temperature. The most important limitation in increasing the temperature of transformers is the drawn current of the transformer. Therefore, by thermal modeling of the transformer, the hot spot temperature can be predicted and a wide view can be given to the operator for the long-term performance of the transformer and its loading current limit. This paper proposes a new method to estimate thermal model coefficients of transformers using Unscented Kalman filter (UKF). By using the proposed estimation method, maximum loading current of transformer which may cause failure in transformer can be calculated. To evaluate the effectiveness of proposed method, practical data of an ۸۰۰ kVA transformer is used to model the top-oil temperature of the transformer. Maximum allowable loading current is calculated online during a ۶۰-hour operation using proposed thermal modeling of the transformer.

Kalman filter, Maximum loading current, Top-oil temperature, Transformer thermal model