Abstract
The objective of this paper is to investigate the ability of physics-informedneural networks to learn the magnetic field response as a function of designparameters in the context of a two-dimensional (2-D) magnetostatic problem. Ourapproach is as follows. First, we present a functional whose minimization isequivalent to solving parametric magnetostatic problems. Subsequently, we use adeep neural network (DNN) to represent the magnetic field as a function ofspace and parameters that describe geometric features and operating points. Wetrain the DNN by minimizing the physics-informed functional using stochasticgradient descent. Lastly, we demonstrate our approach on a\mbox{ten-dimensional} EI-core electromagnet problem with parameterizedgeometry. We evaluate the accuracy of the DNN by comparing its predictions tothose of finite element analysis.