Power Transformers Diagnostics using Heat Model Parameter Identification

Abstract

Condition evaluation of the electrical equipment in operation is an important prerequisite for building reliable and efficient electrical systems and networks. In order to increase accuracy of power transformers condition monitoring, method for analyzing the diagnostic parameters of power transformer oil temperature have been developed. The method is based on parametric identification of the thermal model using particle swarm optimization. In order to test the efficiency of the proposed method, simulation was performed using the experimental data obtained on the power three-winding transformer of the network substation. In view of the impossibility of deliberately changing the technical condition of the transformer, the performance evaluation is based on the confirmation of the absence of changes in the technical state. The analysis of the obtained results confirms the efficiency of the developed method and algorithms.