Particle Swarm Optimization of a Multi-Coil Transverse Flux Induction Heating System

In this paper the optimization of a transverse flux induction heating device by means of a parameter-free, self-adaptive particle swarm optimizer is presented. The novelty of this work is represented by the fact that concentric multi-coil induction heaters have never been optimized before in their full three-dimensional complexity. The main technical challenge of the problem consists in designing a single configuration capable of heating strips of different widths while satisfying geometrical, electrical and heating process constraints. It will be shown that a system consisting of three coils can successfully heat strips of widths between 600 mm and 800 mm with the required uniformity.