Introduction of Artificial Neural Networks in EMC

Abstract

Electromagnetic examinations are usually very expensive. Every simulation needs time for computation and every measurement needs time for preparation. In addition, similar results are generally expected when examining similar objects. If the relation between the differences of investigated objects and their results could be found, a prediction on objects which are not yet examined would be possible. In this paper, a method based on artificial neural networks will be presented, with which a prediction of simulation results of similar objects is possible.