Premium
A bisection‐based heuristic for rapid EM‐driven multiobjective design of compact impedance transformers
Author(s) -
Bekasiewicz Adrian,
Koziel Slawomir
Publication year - 2018
Publication title -
international journal of numerical modelling: electronic networks, devices and fields
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.249
H-Index - 30
eISSN - 1099-1204
pISSN - 0894-3370
DOI - 10.1002/jnm.2523
Subject(s) - multi objective optimization , mathematical optimization , pareto principle , transformer , computer science , simulated annealing , electrical impedance , bisection method , heuristic , fidelity , algorithm , mathematics , engineering , electrical engineering , voltage , telecommunications
Design of microwave structures is a multiobjective task where several conflicting requirements have to be considered at the same time. For contemporary circuits characterized by complex geometries, multiobjective optimization cannot be performed using standard population‐based algorithms due to high cost of electromagnetic (EM) evaluations. In this work, we propose a deterministic approach for fast EM‐driven multiobjective design of microwave structures. Our Pareto ranking bisection algorithm (PRBA) generates candidate designs by dividing the line segments connecting previously obtained Pareto optimal solutions and refining them by means of poll‐type search involving Pareto ranking. Computational efficiency of the optimization process is ensured by a small number of objective function evaluations required by PRBA as well as by executing the algorithm using the low‐fidelity EM model of the structure at hand. The algorithm is demonstrated using a compact impedance transformer. Experimental validation is also provided.