Please note that there are guidelines (opens in new tab) and Latex templates available to help creating presentations, writing reports, Bachelor's or Master's theses.
Thesis topics
Bachelor thesis, Master thesis, Projectseminar, HiWi Position
Parallel algorithms play an increasingly vital role in both research and industry for accelerating simulations. Domain Decomposition methods (DDMs), such as nonoverlapping Schwarz or Dirichlet-Neumann/Neumann-Neumann methods, are effective approaches for introducing spatial concurrency, thereby facilitating parallel computations. When the problems under examination are not only spatially dependent but also time-dependent, waveform relaxation enhances information exchange between different time steps [2, 3]. This, in turn, enhances additional concurrency in time when combined with Parareal methods.
The objective of this project is to implement selected algorithms from the provided references and evaluate their performance using discretized benchmark problems.
Supervisors: Mario Mally, M.Sc., Timon Seibel, M.Sc.
Seminar paper, Bachelor thesis, Proseminar, Projectseminar
In particle accelerators, commonly multi-cell resonating structures are used, e.g. the superconducting 9-cell TESLA cavity. Such devices are manufactured from niobium sheets which are deep-drawn and welded together. During this welding process, misalignment occurs at the joints of the individual components, which leads to deviations from the desired eigenfrequencies and field patters, which negatively influences the accelerating performance of the electromagnetic cavities.
In order to quantify these uncertainties,700 observations are available from measurements at Deutsches Elektronen-Synchrotron (DESY) and found seven uncorrelated deformation variables to describe the cell misalignment.
The purpose of this thesis is to develop an optimization algorithm which allows for conclusions from the measured eigenfrequencies to the deformations of the cavity under consideration. To this end, the closed-form sensitivities with respect to the deformation variables should be used. Using these, an efficient optimization routine should be implemented to identify the misalignment.
Supervisors: Anna Ziegler, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Development of a Stream Function Approach for Accelerator Magnet Design using the Bembel C++ Library
2023
Bachelor thesis
The determination of optimal current distributions on given surfaces is often the starting point for the electromagnetic design of coil dominated electromagnets utilized, e.g. in particle accelerators and magnetic resonance imaging (MRI). Geometrical and mechanical constraints as well as the particle beam size determine the coil winding surface, on which an optimal current density is to be determined.
In this Bachelor thesis, a stream function approach for the determination of optimal current densities on given surfaces shall be developed and implemented in Bembel, the BEM-based engineering library, see www.bembel.eu. The goal is an automated generation of winding paths, on given surfaces and field quality requirements.
Supervisors: Maximilian Nolte, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis
The classical way of expressing the field uniformity in accelerator magnets is by means of the Fourier coefficients of the trigonometric eigenfunctions of the Laplace equation also known as multipoles. When the magnets are curved (Fig. 1), higher-order terms appear and the scaling laws derived for straight magnets are no more applicable. The field reconstruction from field simulations or magnetic measurements should therefore be based on the eigenfunctions of the Laplace equation in the toroidal coordinate system, so called toroidal harmonics.
TU Darmstadt, together with the European Organization for Nuclear Research (CERN) formulated a linear inverse problem to determine the toroidal harmonics from the magnetic flux density.
In this thesis, different algorithms to solve this inverse problem shall be studied and compared on magnetic flux density measurements.
Supervisors: Luisa Fleig, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis, Master thesis, Projectseminar
High-temperature superconducting (HTS) materials are a promising technology for high-field magnets in particle accelerators. In particular, no-insulation (NI) coils, i.e., coils wound without turn-to-turn insulation, have gained popularity due to their robustness [1]. Numerical methods such as the finite element (FE) method play a key role in developing HTS-based applications. The objective of this project is to extend CERN’s existing open-source Finite Element Quench Simulation (FiQuS) framework with 2D axisymmetric FE models of NI coils [2]. Since they are more efficient but less general than existing 3D models, they will complement the latter as an important tool in FiQuS to analyze NI coils. Following FiQuS’s ideals, a key aspect will be to hide the complexities CERN of the FE formulation from the users who are typically not FE experts.
Supervisors: Erik Schnaubelt, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis, Master thesis, Projectseminar
CERN started a project last year to build a comprehensive open-source quench simulation tool called FiQuS (Finite Element Quench Simulation). It is based on the finite element (FE) framework ONELAB and written mostly in Python. The idea is to build a flexible tool which allows users to build and simulate complex models from human-readable inputs hiding the complexity of the FE kernel. In order to simulate real-world accelerator magnet circuits, this project aims at coupling FE magnet models from FiQuS with electric circuits [1], which can then be solved by circuit simulators such as Xyce. A key aspect will be to hide the complexities of this coupling from the users who are typically not FE experts.
Supervisors: Erik Schnaubelt, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Master thesis
Solving partial differential equations numerically with boundary element methods requires the application of fast methods to be competitive with other numerical methods. At high-frequencies, the existing implementation of H2-matrices breaks down and needs to be adapted in order to work efficiently.
The idea of the approach is to approximate a spherical wave in far distance hierarchically by plane waves. The implementation is carried out in the C++ library Bembel, see www.bembel.eu.
Supervisors: Maximilian Nolte, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis, Master thesis, Projectseminar
Electric energy conversion is a key issue on the way to decarbonization. Computational design and optimization of electric motors is a very active research area with the aim to increase the efficiency and power density of electric drives. Yet, optimization in commercial solvers is often performed using time-consuming methods such as surrogates or genetic algorithms, taking days or weeks for one optimization.
This work combines the modeling of the motor using Isogeometric Analysis (IGA), which allows to exactly represent the geometry, with fast gradient based optimization. By using present state-of-the-art numerical modeling techniques together with efficient optimization algorithms, it is possible to reduce the optimization time to several minutes.
Supervisors: Michael Wiesheu, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis
The need for higher energy efficiency and decarbonization give rise to a steadily increasing importance of electric drives. Simulations allow the physical limits to be pushed in order to increase the power density and make motors more cost-efficient. This work aims to investigate the influence of mechanical stresses in electric motors on the electromagnetic behavior and find out how stress dependent material properties can be mitigated or exploited. Simulations are performed in an Finite Element (FE) framework using Isogeometric Analysis (IGA), which allows to exactly represent the geometry. This enables an efficient coupling of the geometric, magnetic and mechanical systems.
Supervisors: Michael Wiesheu, M.Sc., Prof. Dr. rer. nat. Sebastian Schöps
Seminar paper, Bachelor thesis, Master thesis, Projectseminar
Deviations in the manufacturing process of electronic components may lead to rejections due to malfunctioning. Uncertain design parameters (i.e. geometrical and material parameters) can be modeled as random variables. Then, the failure probability of a realization can be estimated. A standard approach for estimating failure probabilities is a Monte Carlo analysis. In a Monte Carlo analysis a large number of sample points is generated according to a given probability distribution. The percentage of sample points not fulfilling some predefined performance feature specifications denotes the failure probability. In order to obtain a reliable estimation, a large number of sample points is required. This leads to high computing costs, since for each sample point a PDE must be solved, e.g. with the finite element method (FEM). Current research deals with the reduction of computational effort. Importance sampling is an approach to reduce the number of FEM evaluations by generating sample points in critical regions with a higher probability.
Supervisors: Dr.-Ing. Mona Fuhrländer, Prof. Dr. rer. nat. Sebastian Schöps
Bachelor thesis, Master thesis
Due to the growing importance of e-mobility, the efficient simulation and optimization of electric energy converters, in particular electric machines, is becoming increasingly important. In the manufacturing process of these electric machines imperfections and small deviations from the nominal design can occur. In the worst case, these imperfections can lead to a significant decrease in quality or even failure of the machine. To avoid this, the machine can be optimized robustly, i.e., considering the deviations in the machine design. Robust optimization aims to find a machine design which is robust in terms of deviations from the nominal design, i.e., to find an optimum in terms of a goal function J which does not deteriorate significantly for small changes in the design parameters p, compare Fig. 2. In this project an electric machine is simulated using isogeometric analysis and the shape of the machine shall be optimized robustly considering uncertainties, using uncertainty quantification methods like the Monte Carlo method.
Supervisors: Dr.-Ing. Melina Merkel, Prof. Dr. rer. nat. Sebastian Schöps
