Applied Design Automation for Finding Feasible Designs for High-Frequency Planar Transformers

Abstract:
The development of planar transformers for high frequency applications is a multi-physics design problem. Various parasitic effects and thermal limitations must be considered. Parasitic elements behave in opposite ways, so a trade-off must be found to meet the design requirements. During the design process, the designs are optimized iteratively. For this purpose, several designs are manually generated and analyzed by electric field simulation tools. However, this optimization is a time-consuming procedure with repetitive tasks. Design automation methods can be used to automate and, thus, to accelerate this optimization process. This paper applies design automation to identify feasible designs for high frequency planar transformers. Automated 2D electric field simulations in conjunction with an evolutionary algorithm are utilized to optimize the designs. The implemented algorithm can perform complex operations to make substantial changes in the transformer layout in order to consider design improvement techniques during the optimization process. Selected prototypes are assembled, electrical characterized and compared with the simulation results to validate the algorithm applied.