%0 Thesis %1 dimcic2012structural %A Dimcic, Milos %B Forschungsberichte aus dem Institut für Tragkonstruktionen und Konstruktives Entwerfen %C ITKE, Stuttgart, Germany %D 2012 %E Knippers, Jan %I Stuttgart: Institut für Tragkonstruktionen und Konstruktives Entwerfen %K 2012 algorithm architecture based dimcic diss dissertation engineering forschungsbericht from:petraheim genetic grid itke optimisation optimization shell structural %T Structural Optimization of Grid Shells based on Genetic Algorithms %V 32 %X In the 21st century, as free form design gains popularity, free-form grid shells are becoming a universal structural solution, enabling merger of structure and facade into a single layer - a skin 31. The subject of the presented work is the optimization of grid structures over some predefined free form shape, with the goal of generating a stable and statically efficient structure. It is shown how combining design and FEM software in an iterative, Genetic Algorithms based, optimization process, stress and displacements in grid shell structures can be significantly reduced, whereby material can be saved and stability enhanced. Within this research, design and static analysis software are combined in order to perform a statical optimization of grid shells,generated over a given free form surface. A plug-in for Rhinoceros 3D (software based on NURBS 44 geometry representation) is developed, that uses Genetic Algorithms as an optimization method and implements automated iterative calls to Oasys GSA (commercial FEM static analysis software) in order to generate a statically optimal grid shell. To make this possible, within this research some new types of automatic grid generation are developed. Voronoi diagrams 11 were used together with the adapted Force-Density method 38 to develop a new type of grid structure that we called Voronax. In the presented work it was shown that, using the same free form surface, and using the same number of joints and structural members, we can generate much more efficient grid shells, when compared to the standard (uniform) grid structures, simply by modifying the structural grid, i.e., rearranging the structural members of the grid shell. The work presented offers an explanation of the entire method and how it can be constructed. The results of the experiments are there to prove its efficiency and credibility. Once it is proved that the method works, its application can take various forms and be left to the creativity of the user and the requirements of the specific project. %@ 978-3-922302-32-2

@phdthesis{dimcic2012structural, abstract = {In the 21st century, as free form design gains popularity, free-form grid shells are becoming a universal structural solution, enabling merger of structure and facade into a single layer - a skin [31]. The subject of the presented work is the optimization of grid structures over some predefined free form shape, with the goal of generating a stable and statically efficient structure. It is shown how combining design and FEM software in an iterative, Genetic Algorithms based, optimization process, stress and displacements in grid shell structures can be significantly reduced, whereby material can be saved and stability enhanced. Within this research, design and static analysis software are combined in order to perform a statical optimization of grid shells,generated over a given free form surface. A plug-in for Rhinoceros 3D (software based on NURBS [44] geometry representation) is developed, that uses Genetic Algorithms as an optimization method and implements automated iterative calls to Oasys GSA (commercial FEM static analysis software) in order to generate a statically optimal grid shell. To make this possible, within this research some new types of automatic grid generation are developed. Voronoi diagrams [11] were used together with the adapted Force-Density method [38] to develop a new type of grid structure that we called Voronax. In the presented work it was shown that, using the same free form surface, and using the same number of joints and structural members, we can generate much more efficient grid shells, when compared to the standard (uniform) grid structures, simply by modifying the structural grid, i.e., rearranging the structural members of the grid shell. The work presented offers an explanation of the entire method and how it can be constructed. The results of the experiments are there to prove its efficiency and credibility. Once it is proved that the method works, its application can take various forms and be left to the creativity of the user and the requirements of the specific project.}, added-at = {2020-06-24T13:42:52.000+0200}, address = {ITKE, Stuttgart, Germany}, author = {Dimcic, Milos}, biburl = {https://puma.ub.uni-stuttgart.de/bibtex/24227971b3516b084125d1b37e8dcbac0/petraheim}, editor = {Knippers, Jan}, howpublished = {Dissertation}, institution = {Institut für Tragkonstruktionen und Konstruktives Entwerfen}, interhash = {62f2af1051cae336f6d6ec10e72fed49}, intrahash = {4227971b3516b084125d1b37e8dcbac0}, isbn = {978-3-922302-32-2}, keywords = {2012 algorithm architecture based dimcic diss dissertation engineering forschungsbericht from:petraheim genetic grid itke optimisation optimization shell structural}, language = {eng}, publisher = {Stuttgart: Institut für Tragkonstruktionen und Konstruktives Entwerfen}, school = {Universität Stuttgart}, series = {Forschungsberichte aus dem Institut für Tragkonstruktionen und Konstruktives Entwerfen}, timestamp = {2021-07-01T13:01:33.000+0200}, title = {Structural Optimization of Grid Shells based on Genetic Algorithms}, volume = 32, year = 2012 }