Inverse Optimization of Core Shape of Electromagnetic Devices using Genetic Algorithms

In this paper a new method for the inverse optimization of core shapes of electromagnetic devices using Genetic Algorithms is presented. For the generation of an initial population and its offspring, the available core area is directly encoded using binary technique. Two methods for selection of genes are proposed: non-restricted method and restricted method. Both methods are applied to the optimization of a metal mold cavity inside permanent magnet production equipment. Electromagnetic analysis is performed using the 2-D finite element method. The robustness of the Genetic Algorithm and the accuracy of the obtained results are very promising