An improved teaching‐learning–based optimization for design of frequency reconfigurable antennas

Abstract

SummaryThis paper proposes an improved teaching‐learning–based optimization (ITLBO) for the electromagnetics and antenna community. The basic TLBO was applied to many different problems with consistent success. For improving the algorithm performance, in the learner phase, each group of two learners is replaced by three‐member groups. Furthermore, two new phases are added to the algorithm. The first phase involves selection of several better learners for teaching subjects to randomly selected weaker learners. The second phase is a mutation operation. The presented algorithm with these changes simulates a classroom more realistically and avoids being trapped in local optima. First, the ITLBO is evaluated to optimize seven benchmark functions and is shown to be faster and better. In addition, a frequency reconfigurable antenna design is considered as a practical problem and Taguchi's method is applied for adjusting the control parameters of the algorithm. Comparisons of the performance of this algorithm with those of the TLBO, two other improved versions of TLBO (I‐TLBO and improved TLBO with learning experience of other learners [LETLBO]) and genetic algorithms demonstrate the higher efficiency for this algorithm in electromagnetic problems. A prototype of the proposed antenna is fabricated, and the simulation and measurement results agree well.