TU Darmstadt / ULB / TUbiblio

Generic Topology Optimization Based on Local State Features

Aulig, Nikola (2017)
Generic Topology Optimization Based on Local State Features.
Book, Primary publication

Abstract

The automatic creation of optimal concepts for mechanical structures in the computer-aided design process has become an important area of research. Continuum topology optimization methods determine the distribution of material within a pre-defined design space and, thus, not only the shape, but also the fundamental geometric layout of a structure. For this task, the majority of the existing, numerical optimization methods requires mathematical gradient information. However, when addressing optimization problems that involve highly non-linear or black-box simulations, it can be difficult to obtain satisfactory results or gradient information at all. In order to provide design concepts also for these types of problems, this thesis presents a generic topology optimization approach. The novel method realizes a self-contained learning component that utilizes physical simulation data to generate a search direction. Based on a continuous problem formulation, every design variable is improved iteratively by a learned update-signal. The individual update-signals are computed from local state features and substitute sensitivities of the design variables. Evolutionary optimization or supervised learning adapt the model parameters for determination of the update-signals to the chosen optimization goal. In empirical studies, the novel method reproduces reference structures with minimum compliance. When applied to a practical problem from the challenging domain of vehicle crashworthiness optimization, specifically the minimization of intrusion, it provides superior design concepts when compared to a frequently applied heuristic method. The results confirm that the proposed method is capable to yield innovative solutions to so far unsolved topology optimization problems.

Item Type: Book
Erschienen: 2017
Creators: Aulig, Nikola
Type of entry: Primary publication
Title: Generic Topology Optimization Based on Local State Features
Language: German
Referees: Adamy, Prof. Dr. Jürgen ; Sendhoff, Prof. Dr. Bernhard
Date: 2017
Place of Publication: Düsseldorf
Publisher: VDI Verlag
Series: Fortschritt-Berichte VDI : Reihe 20
Series Volume: 468
Refereed: 4 May 2017
URL / URN: http://tuprints.ulb.tu-darmstadt.de/6861
Abstract:

The automatic creation of optimal concepts for mechanical structures in the computer-aided design process has become an important area of research. Continuum topology optimization methods determine the distribution of material within a pre-defined design space and, thus, not only the shape, but also the fundamental geometric layout of a structure. For this task, the majority of the existing, numerical optimization methods requires mathematical gradient information. However, when addressing optimization problems that involve highly non-linear or black-box simulations, it can be difficult to obtain satisfactory results or gradient information at all. In order to provide design concepts also for these types of problems, this thesis presents a generic topology optimization approach. The novel method realizes a self-contained learning component that utilizes physical simulation data to generate a search direction. Based on a continuous problem formulation, every design variable is improved iteratively by a learned update-signal. The individual update-signals are computed from local state features and substitute sensitivities of the design variables. Evolutionary optimization or supervised learning adapt the model parameters for determination of the update-signals to the chosen optimization goal. In empirical studies, the novel method reproduces reference structures with minimum compliance. When applied to a practical problem from the challenging domain of vehicle crashworthiness optimization, specifically the minimization of intrusion, it provides superior design concepts when compared to a frequently applied heuristic method. The results confirm that the proposed method is capable to yield innovative solutions to so far unsolved topology optimization problems.

Alternative Abstract:
Alternative abstract Language

Die automatische Erstellung von optimalen Entwurfskonzepten für mechanische Strukturen im rechnergestützten Entwicklungsprozess ist ein wichtiger Forschungszweig. Methoden der Topologieoptimierung bestimmen die Materialverteilung in einem vordefinierten Entwurfsraum und daher nicht nur die Form, sondern auch die grundsätzliche geometrische Ausgestaltung einer Struktur. Die Mehrheit der verfügbaren numerischen Optimierungsmethoden benötigen hierfür mathematische Gradienteninformation. Betrachtet man jedoch Optimierungsprobleme, die stark nichtlineare oder Blackbox-Simulationen beinhalten, kann es schwierig sein, zufriedenstellende Ergebnisse oder überhaupt Gradienteninformation zu erhalten. Um auch für solche Probleme Entwurfskonzepte zu finden, wird in dieser Dissertation ein generischer Topologieoptimierungsansatz präsentiert. Die neue Methode realisiert eine eigenständige Lernkomponente, welche in der Lage ist, aus physikalischen Simulationsdaten eine Suchrichtung zu erstellen. Basierend auf einer kontinuierlichen Formulierung des Problems wird jede Entwurfsvariable durch ein gelerntes Updatesignal iterativ verbessert. Die individuellen Updatesignale berechnen sich aus lokalen Zustandsmerkmalen und ersetzen die Sensitivitäten der Entwurfsvariablen. Evolutionäre Optimierung oder überwachte Lernverfahren passen die Modellparameter zur Bestimmung der Updatesignale an das gewählte Optimierungsziel an. In empirischen Studien reproduziert die neue Methode Referenzstrukturen mit minimaler Nachgiebigkeit. Bei der Anwendung auf ein Problem aus dem anspruchsvollen Gebiet der Optimierung des Fahrzeug-Unfallverhaltens, speziell der Minimierung der Eindringtiefe, liefert sie überlegene Entwurfsvorschläge im Vergleich mit einer häufig verwendeten heuristischen Methode. Die Ergebnisse bestätigen, dass die vorgeschlagene Methode in der Lage ist, innovative Lösungen für bisher ungelöste Topologieoptimierungsprobleme zu erzeugen.

German
URN: urn:nbn:de:tuda-tuprints-68616
Additional Information:

Fortschr.-Ber. VDI Reihe 20 Nr. 468, ISSN 0178-9473

Classification DDC: 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau
Divisions: 18 Department of Electrical Engineering and Information Technology
18 Department of Electrical Engineering and Information Technology > Institut für Automatisierungstechnik und Mechatronik
18 Department of Electrical Engineering and Information Technology > Institut für Automatisierungstechnik und Mechatronik > Control Methods and Robotics (from 01.08.2022 renamed Control Methods and Intelligent Systems)
Date Deposited: 29 Oct 2017 20:55
Last Modified: 29 Oct 2017 20:55
PPN:
Referees: Adamy, Prof. Dr. Jürgen ; Sendhoff, Prof. Dr. Bernhard
Refereed / Verteidigung / mdl. Prüfung: 4 May 2017
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details