Multiobjective design of transverse flux induction heating device for semiconductor processing

A multi-objective optimization of an induction device used for heating thin flat semiconductor plates is performed. Two conflicting targets are considered: increase of the uniformity of dissipated power (and thus of the temperature) and the increase of total power absorbed by the semiconductor. Three optimization techniques, mono-objective aggregation, Deb's algorithm and cluster groups, all implemented using genetic algorithms, are used.