The oil temperature of the oil-immersed power transformer is a key factor. It has a great influence on the transformer operation life and load capacity

and a close relationship with the top oil layer temperature in transformer. Transformers in service at high temperature and overload condition are easy to cause early failure resulted from overheating of winding hot-spot temperature

and the top oil temperature is an important criteria of transformer hot-spot temperature. It is proposed a differential equation

in order to predict accurately the transformer top oil temperature

for predicting the top oil temperature after taking the thermal dynamic effect of load and the environment temperature variations into account

and employed the Kalman filter method to build a real time estimation model for top oil temperature based on state equations and measurement equations. At last

the comparative analysis to the prediction accuracy of Kalman filter model and IEEE guidelines recommend model combined with online monitored values is made. The results show that the Kalman filter model has more prediction accuracy than IEEE guidelines recommend model

and which could be as a new reference to the top oil layer temperature prediction for oil-immersed power transformers.