Integrated structural modelling, adaptive analysis and shape optimization

The present work is concerned with the development of procedures for an integrated system for computer aided structural optimization including topology optimization, image processing and shape optimization for two dimensional structures. At the beginning, such a system is introduced. This system int...

Full description

Bibliographic Details
Main Author: Sienz, J.
Published: Swansea University 1994
Subjects:
Online Access:http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.639033
id ndltd-bl.uk-oai-ethos.bl.uk-639033
record_format oai_dc
spelling ndltd-bl.uk-oai-ethos.bl.uk-6390332015-05-02T03:27:42ZIntegrated structural modelling, adaptive analysis and shape optimizationSienz, J.1994The present work is concerned with the development of procedures for an integrated system for computer aided structural optimization including topology optimization, image processing and shape optimization for two dimensional structures. At the beginning, such a system is introduced. This system integrates aspects of design and analysis modelling, behaviour sensitivity analysis, optimization and creative control via interactive graphics. A satisfactory solution of structural optimization problems can only be achieved by having a profound knowledge of each of these aspects. The introduction of an integrated system for structural optimization is followed by the establishment of a pool of <i>blocks</i> to build such a system. Firstly, a flexible shape definition tool for two dimensional structures using <i>B</i>-splines is shown. This is followed by an adaptive, automatic mesh generator for unstructured meshes consisting of triangular or quadrilateral meshes which allows to generate full FE problem definitions. The FE analysis is carried out using this input for plane stress/plane strain problems, solids of revolution and for engineering problems governed by the Helmholtz equations. The accuracy of the FE analysis is improved by applying adaptive refinement methods in combination with the Zienkiewicz-Zhu error estimator and nodal averaging or the SPR method to compute nodal stresses. For shape optimization, the behaviour sensitivities of the structure are computed using semi-analytical methods (<i>exact</i> or conventional) or the global finite difference method. The optimization is carried out using mathematical programs which are used as <i>black boxes</i>. These tools are then combined to build two applications for integrated structural optimization.502.85Swansea University http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.639033Electronic Thesis or Dissertation
collection NDLTD
sources NDLTD
topic 502.85
spellingShingle 502.85
Sienz, J.
Integrated structural modelling, adaptive analysis and shape optimization
description The present work is concerned with the development of procedures for an integrated system for computer aided structural optimization including topology optimization, image processing and shape optimization for two dimensional structures. At the beginning, such a system is introduced. This system integrates aspects of design and analysis modelling, behaviour sensitivity analysis, optimization and creative control via interactive graphics. A satisfactory solution of structural optimization problems can only be achieved by having a profound knowledge of each of these aspects. The introduction of an integrated system for structural optimization is followed by the establishment of a pool of <i>blocks</i> to build such a system. Firstly, a flexible shape definition tool for two dimensional structures using <i>B</i>-splines is shown. This is followed by an adaptive, automatic mesh generator for unstructured meshes consisting of triangular or quadrilateral meshes which allows to generate full FE problem definitions. The FE analysis is carried out using this input for plane stress/plane strain problems, solids of revolution and for engineering problems governed by the Helmholtz equations. The accuracy of the FE analysis is improved by applying adaptive refinement methods in combination with the Zienkiewicz-Zhu error estimator and nodal averaging or the SPR method to compute nodal stresses. For shape optimization, the behaviour sensitivities of the structure are computed using semi-analytical methods (<i>exact</i> or conventional) or the global finite difference method. The optimization is carried out using mathematical programs which are used as <i>black boxes</i>. These tools are then combined to build two applications for integrated structural optimization.
author Sienz, J.
author_facet Sienz, J.
author_sort Sienz, J.
title Integrated structural modelling, adaptive analysis and shape optimization
title_short Integrated structural modelling, adaptive analysis and shape optimization
title_full Integrated structural modelling, adaptive analysis and shape optimization
title_fullStr Integrated structural modelling, adaptive analysis and shape optimization
title_full_unstemmed Integrated structural modelling, adaptive analysis and shape optimization
title_sort integrated structural modelling, adaptive analysis and shape optimization
publisher Swansea University
publishDate 1994
url http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.639033
work_keys_str_mv AT sienzj integratedstructuralmodellingadaptiveanalysisandshapeoptimization
_version_ 1716802630634700800