Polenz, Björn (2024)
Robust Shape Optimization of Electromechanical Energy Converters.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00028161
Ph.D. Thesis, Primary publication, Publisher's Version
Text
Polenz_Diss.pdf Copyright Information: CC BY-NC-SA 4.0 International - Creative Commons, Attribution NonCommercial, ShareAlike. Download (3MB) |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Robust Shape Optimization of Electromechanical Energy Converters | ||||
Language: | English | ||||
Referees: | Ulbrich, Prof. Dr. Stefan ; Schöps, Prof. Dr. Sebastian | ||||
Date: | 7 October 2024 | ||||
Place of Publication: | Darmstadt | ||||
Collation: | x, 167 Seiten | ||||
Date of oral examination: | 27 October 2023 | ||||
DOI: | 10.26083/tuprints-00028161 | ||||
Abstract: | This work deals with the simulation and shape optimization of electromechanical energy converters under uncertainty. More precisely, an asynchronous machine is considered, whose electromagnetic fields can be described by the magnetoquasistatic approximation of Maxwell’s equations, which are coupled with network equations for the rotor cage and for the exciting three-phase current. The state system is completed by an equation of motion which is excited by the torque. This leads to a system of partial differential algebraic equations. A finite element approach with a time-stepping method is used to solve the equation numerically. We consider uncertainties in the material and geometry of the machine and use a worst-case approach to address these uncertainties. This leads to a bi-level structured optimization problem. Since these problems are difficult to solve numerically, we use approximations up to second order as surrogate models. In particular, we use Taylor models in combination with an adaptive strategy to improve the approximation quality and derivative-free interpolation models that can also be improved iteratively. Both the problem formulation and the consideration of uncertainty in the optimization lead to a high computational cost. To speed up our computations, we apply model dimension reduction techniques. |
||||
Alternative Abstract: |
|
||||
Status: | Publisher's Version | ||||
URN: | urn:nbn:de:tuda-tuprints-281612 | ||||
Classification DDC: | 500 Science and mathematics > 510 Mathematics | ||||
Divisions: | 04 Department of Mathematics > Optimization > Nonlinear Optimization | ||||
TU-Projects: | Bund/BMBF|05M18RDA|PASIROM | ||||
Date Deposited: | 07 Oct 2024 13:34 | ||||
Last Modified: | 15 Oct 2024 06:49 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/28161 | ||||
PPN: | 522021700 | ||||
Export: |
View Item |