Export bibliographic data
Literature by the same author
plus on the publication server
plus at Google Scholar


Evolutionary Multi-Objective Topology Optimization for Engineering Problems : M10/17

Title data

Schleifer, Felix:
Evolutionary Multi-Objective Topology Optimization for Engineering Problems : M10/17.
Bayreuth , 2017 . - 86 S.
(Master's, 2017 , Universität Bayreuth, Fakultät für Ingenieurwissenschaften, Lehrstuhl Konstruktionslehre CAD)

Abstract in another language

In engineering optimization is a mathematical strategy aiming at manipulating a model of a real system so that one or more properties and specific values of the system are improved. For the diverse demands of modern engineering a series of different specific optimization methods have been developed. The system response is provided by a computer simulation, so the procedure can be carried out fully automated. Computer-based optimization methods play an integral role in the design of e. g. production processes, air routes, control systems and geometric structures. For the engineering design process specialized methods were developed aiming to find optimal distributions of solid material. An expensive trial and error based development can be avoided by such cost and time efficient structural optimization procedures. Nowadays, especially in the industrial sector of mobility, where lighter structures lead to better performing and less fuel consuming products, the use of computer aided optimization is omnipresent and almost inevitable. The topology optimization itself is a fundamental design step that requires little to no a priori knowledge about the structure, yet only information about the available design space and the expected loading conditions. Multi-objective optimization aims to find a set of solutions to a problem that can not be outperformed with respect to all optimization goals. The set of trade-off solutions that is the result of such an multi-objective optimization can lead to more specific insights about the design problem in comparison to classical approaches. Amongst the approaches to solve a multi-objective optimization problem the evolutionary algorithms have proven to be promising choices. Evolutionary algorithms have already been used for multiobjective topology optimization and have demonstrated the ability to provide the engineer with a set of near-application trade-off solutions. The objective of this thesis is to develop a multi-objective topology optimization method for engineering design. The most promising algorithm from literature has to be chosen and, if necessary, expanded to account for problem-specific demands. Because most examinations have only yet focused on showing that multi-objective structural optimization is possible, in this thesis more near-application problems will be tackled. This includes large structures, non-rectangular design spaces, and three dimensional models. Optimization goals will be up to three independent objectives derived from mechanical and thermal calculations with up to three different constraints. Methods to reduce computational effort and to enhance the usability will be presented and implemented. The found optimization results will be used to derive sets of trade-off design propositions. The found solutions will be discussed with respect to their qualities and the performance of the optimizer. Physical experiments and simulations will be carried out to evaluate, whether the developed optimization tool can reliably find feasible solutions to the proposed problems. The applicability of the presented method in an engineering design process will be discussed.

Further data

Item Type: Master's, Magister, Diploma, or Admission thesis (Master's)
Keywords: FEA; Topologieoptimierung
Institutions of the University: Faculties > Faculty of Engineering Science > Chair Engineering Design and CAD
Faculties > Faculty of Engineering Science > Chair Engineering Design and CAD > Chair Engineering Design and CAD - Univ.-Prof. Dr.-Ing. Frank Rieg
Faculties > Faculty of Engineering Science
Result of work at the UBT: Yes
DDC Subjects: 600 Technology, medicine, applied sciences > 620 Engineering
Date Deposited: 27 Jul 2017 06:27
Last Modified: 27 Jul 2017 06:27
URI: https://eref.uni-bayreuth.de/id/eprint/38953