Computational Magnetics
Many electro-technical devices such as e.g. printed circuit boards, electrical drives and antenna systems can be simulated on the basis of electrical circuits. However, the increasing frequencies and the decreasing size force designers to account for wave propagation effects, eddy-current effects, ferromagnetic saturation and hysteresis. For wave propagation effects and eddy-current effects, the results of stand-alone field simulation can be represented by an order-reduced equivalent model, which is then inserted in the overall circuit model. The representation of field-dependent nonlinearities and hysteresis effects, however, is not straightforward.

2D Simulation of a Transformer
The coupled field and circuit simulation becomes troublesome when a large number of time steps is required. This occurs when e.g. simulating an electrical drive where the machine requires 10 periods of 50 Hz to reach nominal speed whereas the switching of the Insulated Gate Bipolar Transistors in the frequency converter switches at 20 kHz, necessitating a time steps in the order of a microsecond to be used in the simulation. Since the field model consists typically of a few million degrees of freedom, all those unknowns have to be solved in every time step. Fortunately, the relevant time constants in electrical-energy converter are in the range 50 Hz. Hence the field model does not have to be time-stepped at the same rate as the circuit model, in which fast switches are present. The use of adaptive multirate time-integration schemes can reduce the numerical complexity of the problem substantially.
Research Questions
- Efficiency of the time-integration for field devices in pulsed circuits (multirate, dynamic iteration)
- DAE-index of the coupled system
- Existence and Uniqueness of the solution
Cooperation
- Herbert De Gersem, Katholieke Universiteit Leuven
- Markus Clemens, Bergische Universität Wuppertal
- Sascha Baumanns, Universität zu Köln
Former and ongoing projects
Publications
- 2024
5228.
M., Günther; Jacob, B.; Totzeck, C.
Data-driven adjoint-based calibration of port-Hamiltonian systems in time domain
Math. Control Signals Syst., 36 :957–977
20245227.
Zaspel, Peter; Günther, Michael
Data-driven identification of port-Hamiltonian DAE systems by Gaussian processes
Preprint
20245226.
Allmendinger, Richard; Fonseca, Carlos M.; Sayin, Serpil; Wiecek, Margaret M.; Stiglmayr, Michael
Multiobjective Optimization on a Budget (Dagstuhl Seminar 23361)
2024
Herausgeber: Schloss Dagstuhl – Leibniz-Zentrum für Informatik5225.
Vinod, Vivin; Zaspel, Peter
Investigating Data Hierarchies in Multifidelity Machine Learning for Excitation Energies
20245224.
Botchev, M. A.; Knizhnerman, L. A.; Schweitzer, M.
Krylov subspace residual and restarting for certain second order differential equations
SIAM J. Sci. Comput., 46 (2) :S223-S253
20245223.
Hastir, Anthony; Jacob, Birgit; Zwart, Hans
Linear-Quadratic optimal control for boundary controlled networks of waves
20245222.
Xu, Zhuo; Tucsnak, Marius
LQR control for a system describing the interaction between a floating solid and the surrounding fluid
Mathematical Control and Related Fields, 14(4) :1477-1500
Dezember 20245221.
Costa, G Morais Rodrigues; Ehrhardt, Matthias
Mathematical analysis and a nonstandard scheme for a model of the immune response against COVID-19
Band 793
Seite 251–270
Herausgeber: AMS Contemporary Mathematics
2024
251–2705220.
Costa, G Morais Rodrigues; Ehrhardt, Matthias
Mathematical analysis and a nonstandard scheme for a model of the immune response against COVID-19
Band 793
Seite 251–270
Herausgeber: AMS Contemporary Mathematics
2024
251–2705219.
Bolten, Matthias; Kilmer, Misha E.; MacLachlan, Scott
Multigrid preconditioning for regularized least-squares problems
SIAM J. Sci. Comput., 46 (5) :s271—s295
2024
ISSN: 1064-82755218.
Schultes, Johanna
Multiobjective optimization of shapes using scalarization techniques
Dissertation
Dissertation
Bergische Universität Wuppertal
20245217.
Bolten, Matthias; Doganay, Onur Tanil; Gottschalk, Hanno; Klamroth, Kathrin
Non-convex shape optimization by dissipative {H}amiltonian flows
Engineering Optimization
20245216.
Zaspel, Peter; Günther, Michael
Data-driven identification of port-Hamiltonian DAE systems by Gaussian processes
Preprint
20245215.
Bolten, M.; Doganay, O. T.; Gottschalk, H.; Klamroth, K.
Non-convex shape optimization by dissipative Hamiltonian flows
Eng. Optim. :1—20
20245214.
Bauß, Julius
On improvements of multi-objective branch and bound
Dissertation
Dissertation
Bergische Universität Wuppertal
20245213.
Abel, Ulrich; Acu, Ana Maria; Heilmann, Margareta; Raşa, Ioan
On some Cauchy problems and positive linear operators
Mediterranean Journal of Mathematics, accepted
20245212.
Lorenz, Jan; Zwerschke, Tom; Schaefers, Kevin
Operator splitting for coupled linear port-Hamiltonian systems
20245211.
Kruse, Thomas; Strack, Philipp
Optimal dynamic control of an epidemic
Operations Research, 72 (3) :1031–1048
2024
Herausgeber: INFORMS5210.
Kruse, Thomas; Strack, Philipp
Optimal dynamic control of an epidemic
Operations Research, 72 (3) :1031–1048
2024
Herausgeber: INFORMS5209.
Vinod, Vivin; Kleinekathöfer, Ulrich; Zaspel, Peter
Optimized multifidelity machine learning for quantum chemistry
Mach. Learn.: Sci. Technol., 5 (1) :015054
20245208.
Frommer, Andreas; Ramirez-Hidalgo, Gustavo; Schweitzer, Marcel; Tsolakis, Manuel
Polynomial preconditioning for the action of the matrix square root and inverse square root
Electron. Trans. Numer. Anal., 60 :381-404
20245207.
Jamil, Hamza
Intrusive and non-intrusive uncertainty quantification methodologies for pyrolysis modeling
Fire Safety Journal, 143 :104060
2024
ISSN: 0379-71125206.
Hosfeld, René; Jacob, Birgit; Schwenninger, Felix; Tucsnak, Marius
Input-to-state stability for bilinear feedback systems
SIAM Journal on Control and Optimization, 62 (3) :1369-1389
20245205.
Schäfers, Kevin; Finkenrath, Jacob; Günther, Michael; Knechtli, Francesco
Hessian-free force-gradient integrators
20245204.
Santos, Daniela Scherer; Klamroth, Kathrin; Martins, Pedro; Paquete, Luís
Ensuring connectedness for the Maximum Quasi-clique and Densest $k$-subgraph problems
2024