The Finite Element Method for Electromagnetic Modeling

During the design of an electromagnetic device, the designer should propose a configuration satisfying the functional needs as far as possible, and at the same time, making it viable from an economic point of view.
Very often, the distribution of the electromagnetic field has a considerable impact on the characteristics to be optimized. Here are some design problems in which a good knowledge of the physical behavior is necessary:
minimization of the reluctant torque in a rotating machine;
obtaining a return force with minimization of Joule losses in an electromagnet;
etc.
The criterion to be maximized or minimized is a scalar quantity relevant to the objective to be reached. It can concern an intrinsic variable of the studied device (a torque, a force, total losses, etc.) or a combination of several measurements of space or time (sinusoidal induction in the air-gap of a rotating machine, trajectory of closing of an electromagnetic contactor, etc.). Moreover, the criterion not always being unique, it is sometimes necessary to use a multicriterion optimization.
[1]
The search for the best performance of a device, in which structural (number of teeth, notch types, etc.), dimensional (dimensions of an air-gap, width of notches, etc.) and physical (type of materials, current density, etc.) parameters intervene, is a difficult problem. This is due, mainly, to the complexity of the phenomena which appear alone or combined: magnetic saturations, eddy currents, three-dimensional effects, relative movements of parts, electric couplings, multi-physical couplings,...