TU Darmstadt / ULB / TUbiblio

Derivative Free Surrogate Optimization for Mixed-Integer Nonlinear Black Box Problems in Engineering

Hemker, Thomas (2010):
Derivative Free Surrogate Optimization for Mixed-Integer Nonlinear Black Box Problems in Engineering.
Darmstadt, TU Darmstadt, [Online-Edition: urn:nbn:de:tuda-tuprints-21626],
[Book]

Abstract

Optimization problems based on black boxes arise in engineering applications every day. Such black boxes typically represent the simulated or experimentally obtained behavior of systems for which almost no internal, structural or analytical knowledge can be provided on a relevant level for the optimization's objective. These resulting non-relaxable mixed-integer nonlinear black box-based optimization problems cannot be carried out efficiently by today's optimization methods. This work provides a new general applicable derivative free optimization approach. The performance of this method will be demonstrated for several benchmark and real world problems from electrical engineering, environmental sciences, and robotics. It will be shown that huge improvements of the optimization's objectives can be achieved for all applications, simply by applying a reasonable number of black box evaluations.

Item Type: Book
Erschienen: 2010
Creators: Hemker, Thomas
Title: Derivative Free Surrogate Optimization for Mixed-Integer Nonlinear Black Box Problems in Engineering
Language: English
Abstract:

Optimization problems based on black boxes arise in engineering applications every day. Such black boxes typically represent the simulated or experimentally obtained behavior of systems for which almost no internal, structural or analytical knowledge can be provided on a relevant level for the optimization's objective. These resulting non-relaxable mixed-integer nonlinear black box-based optimization problems cannot be carried out efficiently by today's optimization methods. This work provides a new general applicable derivative free optimization approach. The performance of this method will be demonstrated for several benchmark and real world problems from electrical engineering, environmental sciences, and robotics. It will be shown that huge improvements of the optimization's objectives can be achieved for all applications, simply by applying a reasonable number of black box evaluations.

Place of Publication: Darmstadt
Publisher: TU Darmstadt
Uncontrolled Keywords: Black Box-Based Optimization, Simulation-Based Optimization, Surrogate Functions, Surrogate Models , Derivative Free Optimization, Derivative free Mixed-Integer Nonlinear Programming , Sequential Designs
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Simulation, Systems Optimization and Robotics Group
Zentrale Einrichtungen
Exzellenzinitiative
Exzellenzinitiative > Graduate Schools > Graduate School of Computational Engineering (CE)
Exzellenzinitiative > Graduate Schools
Date Deposited: 26 May 2010 08:43
Official URL: urn:nbn:de:tuda-tuprints-21626
Additional Information:

Druckausgabe: Düsseldorf, VDI-Verl., 2009. ISBN 978-3-18-379710-3 (Fortschritt-Berichte VDI, R. 10, Bd. 797) [Darmstadt, TU, Diss. 2008]

License: Creative Commons: Attribution-Noncommercial-No Derivative Works 3.0
Referees: von Stryk, Prof. Dr. Oskar and den Hertog, Prof. Dr. Dick
Refereed / Verteidigung / mdl. Prüfung: 19 December 2008
Related URLs:
Alternative Abstract:
Alternative abstract Language
Im Rahmen dieser Arbeit werden Fragestellungen betrachtet, in die neben einer expliziten analytischen Formulierung auch das Verhalten von nicht einsehbaren abgeschlossenen Systemen (Black Box) wie komplexen ingenieurwissenschaftlichen Simulationsprogrammen oder reale Roboter (z.B. Roboter) eingeht. Neben den einfließenden Systemzuständen die immer gewissen Störungen unterliegen muss weiterhin muss angenommen werden, dass die Optimierungsvariablen sowohl kontinuierlich als auch rein diskret sein können. Für die entstehenden gemischt ganzzahligen, Black Box basierten, nichtlinearen Optimierungsprobleme existieren noch sehr wenige allgemein anwendbare und effiziente Optimierungsverfahren. Nach einer detaillierten Definition des Problems wird ein Verfahren basierend auf Ersatzfunktionen eingeführt, dessen Effizienz für die beschriebene Problemklasse anhand verschiedener realer Anwendungsbeispiele gezeigt wird.German
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)

View Item View Item