Using Data Analytics to Provide Better Understanding on Technical and Economic Constraints of Owning Power Transformers (261)
The utilities are continually facing regulatory pressure to reduce operational costs while maintaining the historically high network reliability. Consequently, a project was initiated by the University of Queensland, funded by the Australian Research Council and several partner utilities, to develop the technologies and methodologies to determine the most cost effective utilisation for a power transformer. This is significant because a very large proportion of power transformers around the world is reaching, or have passed, their design life. As an example, in Australia the mean age of the power transformer fleet is approaching 30 years compared to a design life of 40 years.
Five research topics have been pursued for this project. One is to improve the modelling of water throughout the insulation structure of the transformer, to improve diagnostic techniques, and to develop fibre optic based sensors to measure water directly adjacent to the winding. Two is to investigate the long term behaviour of biodegradable oils, which although have been around for twenty years there is still insufficient materials on how a transformer filled with such fluids behaves during its full life cycle. Three is to improve the understanding of the relationship between the loss of winding clamping pressure and the pressboard mechanical properties. Four is to advance diagnostic techniques to detect problems in on-load tap changers and bushings, which are still problems for the industry. Five is to develop economic models to determine when it is better value to retire a transformer rather than incur the potential cost of a failure. The development of these activities first took place in the laboratory, and then expanded into field trials conducted with the partner utilities.
The results of these activities will be discussed in this article.