Future Trends in Optimal Design in Electromagnetics
Paolo Di Barba
Abstract
In computational electromagnetics, there are manyfold advantages when using machine learning methods because no mathematical formulation is required to solve the direct problem for given input geometry. Moreover, due to the inherent bidirectionality of a convolutional neural network, it can be trained to identify the geometry giving rise to the prescribed output field. All this puts the ground for neural meta-modeling of fields, despite different levels of cost and accuracy. For the sake of an example, a surrogate model of the field in a small device is shown. In particular, a concept of multi-fidelity model makes it possible to control both prediction accuracy and computational cost. Moreover, TEAM Problem 35 is solved and it is shown how a generative adversarial network can help multiobjective optimal design.