Differential evolution as applied to electromagnetics: Advances, comparisons, and applications

Several optimization strategies have been proposed in the last decades for the solution of problems in the framework of electromagnetics. Because of the high non-linearity and non-convexity of many problems at hand which are characterized by a large number of unknown parameters, global iterative approaches based on stochastic, multiple-agent algorithms have shown working effectively and being able to recover suitable solutions. Among them, the Differential Evolution (DE) has demonstrated being an effective and stable optimization approach, particularly efficient when dealing with continuous spaces. In this paper, the latest advancements on the use of DE-based approaches to the solution of optimization problems in electromagnetics are shown and reviewed. Comparisons with other global optimization approaches will be also reported pointing out advantages and limitations of the different strategies in several fields of application.