Publications Repository - Gdańsk University of Technology

Majority of practical engineering design problems require simultaneous handling of several criteria. Although many of design tasks can be turned into single-objective problems using sufficient formulations, in some situations, acquiring comprehensive knowledge about possible trade-offs between conflicting objectives may be necessary. This calls for multi-objective optimization that aims at identifying a set of alternative, Pareto-optimal designs. The most popular solution approaches to genuine multi-objective optimization include population-based metaheuristics. Unfortunately, such methods are not practical for problems involving expensive computational models, particularly for antenna engineering, where reliable design requires CPU-intensive electromagnetic (EM) analysis. In this work, we discuss two methodologies for expedited multi-objective design optimization of a six-parameter dielectric resonator antenna (DRA) with respect to three design criteria. The considered solution approaches rely on surrogate-based optimization (SBO) paradigm, where the design speedup is obtained by shifting the optimization burden into a cheap replacement model referred to as a surrogate. The latter is utilized for generating the initial approximation of the Pareto front representation as well as further refinement of the initially obtained Pareto-optimal solutions.

Authors

• Adrian Bekasiewicz,
• prof. dr inż. Sławomir Kozieł link open in new tab ,
• prof. dr hab. inż. Włodzimierz Zieniutycz link open in new tab ,
• Leifur Leifsson