Particle Swarm Optimization (PSO) in Engineering Electromagnetics: A Nature-Inspired Evolutionary Algorithm

Abstract

This paper presents representative electromagnetic engineering applications of a particle swarm optimization (PSO) engine developed at UCLA antenna lab. The paper provides a conceptual overview of the PSO engine that accommodates real-number, binary, single-objective and multi-objective optimizations, along with five examples illustrated to validate the algorithm's functionality in a large variety of practical problems. Examples include the design of dual/multi-band patch antennas, aperiodic antenna arrays, periodic structures, correlator antenna arrays, etc. In most cases the designs provided by PSO are prototyped, with high-quality performance observed from the measurement results.