Temperature Measuring-Based Decision-Making Prognostic Approach in Electric Power Transformers Winding Failures

The electric power transformer is a vital apparatus in power systems, and failure prognostics is significant for the protection of this asset. In addition to the asset damage, its unexpected failure would interrupt power delivery and jeopardize the stability of the system. There are several fault diagnosis methods introduced for the detection of this kind of fault; however, their functionality is for the postfault condition when the asset is already damaged, and the operation of the system is interrupted. Electric insulation deteriorations make the transformers susceptive to faults due to thermal and electrical stresses. In this article, the impact of early stages insulation deteriorations on the temperature inside the transformer is studied using a finite-element electromagnetic–thermofluid method and based on the observations an online sensor-based decision-making predictive fault diagnosis approach is proposed. Finally, the results are experimentally verified.