Boundary Condition Independent Thermal Network Modeling of High-Frequency Power Transformers

With increasing global power demands and the trend of miniaturization of power converters, electronic components are being pushed toward their thermal limits. This has accentuated the need for accurate and low-cost thermal models of electronic components. Lumped Parameter Thermal Network Modeling (LPTNM) is commonly used to evaluate the thermal performance of electronic components, but unlike semiconductors, transformer LPTNM is not standardized in the academic community. Due to this lack of standardization, most proposed transformer thermal models have not been evaluated for Boundary Condition Independence (BCI). This study presents a BCI thermal modeling approach for power transformers based on LPTNM. A thermal network model is developed by discretizing the transformer geometry into smaller volumes and evaluating the thermal resistances by approximating the three-dimensional heat transfer problem as a piecewise one-dimensional model. The network model/geometry discretization is iteratively improved for a set of different boundary conditions, resulting in a BCI model. The modeling approach is experimentally and numerically validated on a PQ40/30 transformer operating in a 3.3 kW switch mode power supply. The BCI transformer thermal model closely matches both numerical and experimental results with a fraction of the computational cost of the numerical model. The proposed BCI modeling approach provides a low-cost alternative to detailed thermal models and can predict transformer thermal performance for varied applications.