Field Validated Dynamic Thermal Model for Power Transformer Insulation System Assessment

Mastering the thermal performances of power transformers is an aspect of first importance for the utilities. Because it's the knowledge of the life expectancy or the condition of the transformer. Oil temperature and winding hot spot are the main parameters which have an impact on the transformer condition. Therefore a reliable thermal model represents a tool of a great support to the transformer user. The use of the thermal models is a subject which always raises discussions on their validity. Our experiment is based on an investigation program regarding the thermal model validation using the data recorded over several months in the condition of industrial operation of two GSU transformers. A set of sensors was installed in the transformers having undergoes FAT. These two units of EDF transformers flotilla were in operation in experiencing various stages of the cooling system impacting consequently the oil temperatures and winding hot spot variations. The application of the data recorded on line during the two GSU operation with consideration of the number of operating heat exchanger, pumps and fans, allows to evaluate the models suggested in the IEC and IEEE standards as well as other models. To assess the oil temperature and hot spot of the transformers, a thermal model has been adapted in order to be used as a tool for the transformer monitoring. In this paper we will expose a methodology used for the improvement of the thermal model. In this paper, we will demonstrate how such validated thermal model will be used in the monitoring of other transformer in operation

[1]  Russell C. Eberhart,et al.  A new optimizer using particle swarm theory , 1995, MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science.