The proper orthogonal decomposition method and Arnoldi-based Krylov projection method are investigated in order to reduce a finite-element model of a quasi-static problem. Both methods are compared on an academic example in terms of computation time and precision.

In this paper, proper orthogonal decomposition (POD) method is employed to build a reduced-order model from a high-order nonlinear permanent magnet synchronous machine model with multiple inputs. Three parameters are selected as the multiple inputs of the machine. These parameters are terminal current, angle of the terminal current, and rotation angle. To produce the lower-rank system, snapshots or instantaneous system states are projected onto a set of orthonormal basis functions with small dimension. The reduced model is then validated by comparing the vector potential, flux density distribution, and torque results of the original model, which indicates the capability of using the POD method in the multivariable input problems. The developed methodology can be used for fast simulations of the machine.

The analysis of the electrical behavior of a unidirectional (UD) composite material requires a lot of resources and measurement techniques. To overcome this constraint, the real composite material is replaced by a virtual material with same overall behavior. In this paper, an approach to determine setting parameters of the virtual material model is presented in order to accurately reconstruct the electrical behavior of a real UD material in the three spatial directions.

The Proper Orthogonal Decomposition (POD) combined with the (Discrete) Empirical Interpolation Method (DEIM) can be used to reduce the computation time of the solution of a Finite Element (FE) model. However, it can lead to numerical instabilities. To increase the robustness, the POD_DEIM model must be constructed by preserving the structure of the full FE model. In this article, the structure preserving is applied for different potential formulations used to solve electromagnetic problems.

Among the model-order reduction techniques, the proper orthogonal decomposition (POD) has shown its efficiency to solve magnetostatic and magnetoquasistatic problems in the time domain. However, the POD is intrusive in the sense that it requires the extraction of the matrix system of the full model to build the reduced model. To avoid this extraction, nonintrusive approaches like the data-driven (DD) methods enable to approximate the reduced model without the access to the full matrix system. In this paper, the DD-POD method is applied to build a low-dimensional system to solve a magnetostatic problem coupled with electric circuit equations.