In the research group "High Frequency Technology", classical tasks of electrical engineering from the wide field of computation of high-frequency electromagnetic fields are processed with modern numerical and analytical methods on current high-performance computers. The observable shift of the fields of application to ever higher frequencies, together with the likewise constantly increasing expectations on the modeling accuracy, requires a permanent further development of the applied computational programs in order to effectively meet the changed requirements from the computational practice.

Key Points

Numerical calculation of high-frequency electromagnetic fields in the time and frequency domain using volume-discretizing methods such as finite integration theory, finite element and finite volume methods, as well as surface-discretizing methods such as boundary element methods. Special attention is given to practical problems involving complex geometric structures and material distributions for which precise and robust solutions are sought.

  • Efficient set-up and solution of (nonlinear) eigenvalue problems for waveguides (2-D) or resonators (3-D) as well as combinations thereof
  • Determination of the natural frequency distribution of regular and chaotic resonators under the necessary consideration of many natural frequencies (>1000)
  • Development of fast methods for beam dynamics simulation by efficiently solving the time-dependent Vlasov equation
  • Modeling of resonators for circular accelerators (synchrotron)
  • Development of new resonators for linear accelerators with a focus on efficient coupling and decoupling of external or beam-excited electromagnetic fields
  • Application and further development of parametric model order reduction methods for fast evaluation of transfer functions
  • Simulation of electromagnetic wave propagation in tunable anisotropic materials, which can be realized, for example, in classical waveguide arrays with liquid crystal fillings
  • Determination of a large number of eigenvalues from the low-frequency part of the real spectrum of large symmetric (generalized) eigenvalue formulations
  • Provision of practical tasks of microsystems engineering on the abstract system level, which is further processed for downstream mathematical model order reduction
  • Development of model order reduction methods for the extraction of physically interpretable equivalent circuits from three-dimensional passive conductor arrays