Electromagnetic modeling of wireless power transmission in the presence of foreign objects

OData support
Dr. Bilicz Sándor
Department of Broadband Infocommunications and Electromagnetic Theory

Wireless Power Transfer (WPT) has become a seemingly ubiquitous technology in the past decade. The flexibility to transfer electric energy in free space without using any cable or waveguide has made it a popular, intensively researched field. An efficient electromagnetic model for the magnetically coupled resonant WPT has already been developed at the Department of Broadband Infocommunication and Electromagnetic Theory. The model uses a thin wire approximation and Method of Moments (MoM) as the numerical method to keep the number of degrees of freedom low and provide a by orders of magnitude faster solution compared to commercial solvers using Finite Element Method (FEM). The method is, however, limited to homogeneous media and cannot treat foreign objects (e.g, human body) in the vicinity of the WPT system. The present work is aiming at developing an electromagnetic model that takes these foreign objects into account and is compatible and implementable to the already existing MoM-based field computation code. As a general workflow, the original full-wave model is reduced to a quasistatic model. Also some considerations on the foreign objects are applied, i.e, they are assumed to be of canonical shaped (sphere, thin plate) and pure dielectrics. With these conditions, the foreign objects can be treated via the modification of the electrostatic Green’s function. The modification is based on the principle of dielectric mirroring, that provides an analytic solution for the Green’s function, that is a necessary condition for the efficient applicability

of MoM. The MoM-based model is implemented in MATLAB environment and is validated against 3-dimensional full-wave FEM simulation and experimental data as well. Sensitivity analysis has also been performed to highlight those parameters of the model, that influence the behavior of the system the most. The analysis is based on the calculation of the Sobol’ indices as the factors of the sensitivity. The analysis is enhanced by a Polynomial Chaos Expansion (PCE) surrogate model, the reduces the required number of electromagnetic simulations and saves computational time.


Please sign in to download the files of this thesis.