Improving the utilization of transformers requires that the hot-spot temperatures and top-oil temperatures (TOTs) be predicted accurately. Our experimentation with various discretization schemes and models proved that many of the linear and nonlinear semiphysical and nonphysical models we were using to predict transformer TOT were correctly modeling the TOT behavior. Our experience convinced us that noisy input data and the absence of data on significant driving variables, not model deficiencies, were frustrating our attempts to reduce the prediction error further. In this paper, we discuss the body of research that leads us to these conclusions.
ASJC Scopus subject areas
- Control and Systems Engineering
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering