An Open Platform Of Parameterized Shape Optimization Based On CAD/CAE Integration Techniques

As an application of Computer Aided Engineering (CAE), structure shape design optimization aims to find the optimum shape of a domain which minimizes or maximizes the given criterions. The problem has been well studied with its theory and algorithms for decades; however its application into manufacturing and engineering industry is still far from what have been achieved in research works. Commercial software for structural shape optimization is still under development.Some major barriers are not well overcome which may limit using of structure shape design optimization into engineering applications. They are: 1) lack of general and well-suited structure shape description method, which is extremely important for the design variables definition; 2) for the sake of computation of design sensitivity, modification of structure shape is greatly constrained because in the analytical sensitivity analysis, the variance magnitude of design variables are limited; 3) the optimized results are not always applicable and may not be accepted by designers or engineers because the results may often include some unreasonable shape modifications.Computer Aided Design (CAD) has played important role in modern manufacture industry and high-tech development. It provides powerful computer-based tools which assist engineers, architects and other design professionals in their design activities. Designers almost can hardly set up 3D models and 2D drawings of physical components without CAD tools. CAD tools are widely used throughout the engineering process from conceptual design and layout of products, through strength and dynamic analysis of assemblies to definition of manufacturing methods of components.Integrated CAE with CAD will popularize the application of CAE into application and CAD/CAE concurrent engineering has gained more and more attentions within development of Manufacturing Industry and Information Technology. CAE techniques such as structural design optimization can help to improve product quality, decrease developing expense and shorten developing period.The research in the dissertation is to benefit structural shape optimization from CAD parametric geometry modeling by means of integration. From the viewpoint of product design, structural shape optimization can be more applicable into modern Manufacturing and Engineering Industry through integrated within CAD tools. At the same time, structural shape optimization can help engineer/designer be aware of the structure performance according to its shape and make better design. From the viewpoint of structure shape optimization, parametric geometry modeling techniques in CAD provide parameterized shape description and control method for structural shape optimization which is more effective and intuitional. For multi-disciplinary structural design optimization, parametric geometry modeling within professional CAD tool may be the most fitful method which can feed different requests on modeling from different disciplinary.The main content in the research is developing an open platform of parameterized shape optimization based-on CAD/CAE integration technique. Through integration, structural analysis and design optimization are seamless combined with parametric geometry modeling and embed into the CAD system. POSHAPE can provide parameterized shape optimization method for 3D solid structure, spatial shell structure and cell structure of composite material. To realize such a general method, integration is the most essential part. In this platform, integration is realized includes: 1) Integrating structure analysis tool of different disciplinary with structure shape optimization. Structure response from different disciplinary will be studied according structure shape. 2) Integrating finite element modeling with parametric geometry modeling through Boundary Representative Tree (simplified as: B-Rep) used in solid modeling. Finite element model is parameterized to be dynamic regenerated during optimization design steps. 3) Parametric solid modeling is extended to realize parameterized surface modeling under integration between surface model definition and solid model. Parameterized finite element modeling of shell structure is also achieved which is similar to that of solid structure.The research work will be introduced with the following chapters.In chapter 1, the background, classification and history of structure design optimization is introduced. The problems arise in application of structural shape optimization into engineering is discussed. Parametric design technique is described which is the basic theory of the present research. The research works on composite material design is also given in this chapter. The main work within this research is outlined at last.In chapter 2, present research on CAD/CAE integration will be summarized firstly. The reason to integrating CAE into CAD system will be given subsequently. Parametric feature modeling is precondition to the integration and will be outlined nextly. Then available data transfer methods and architectures of CAD/CAE integration system are listed and compared. Lastly, some of important techniques used for structure shape optimization are listed.In chapter 3, development of the platform is described, which includes developing environment and realization of parametric finite element modeling. Some problems exits in the platform are also given. In chapter 4, implementation of parameterized shape optimization for 3D solid structure is illustrated. Numerical examples came from engineering will testify the effectiveness of the present method.In chapter 5, parameterized surface modeling and parameterized finite element modeling for shell structure based on parametric solid model is introduced. Them implementation of parameterized shape optimization for spatial shell structure is illustrated. Numerical examples are given to show the effectiveness of the present method.In chapter 6, parameterized shape optimization method is extended to composite material design. For composite material with periodical cell structure, the material property can be improved by means optimization the shape of cell structure. Homogenization method for computing the composite material property is present. Implementation of parameterized shape optimization for cell structure is outlined.In the last chapter, the main contributions of the dissertation are summarized and the further works are suggested.In appendix A, functionality and system command of POSHAPE are listed.The research work is part of co-direction of doctorate thesis project between Universite de Reims Champagne Ardenne and Dalian University of Technology. The Ph.D. dissertation presentation was held on Dec. 2006 and accessed the doctor degree of Universite de Reims Champagne Ardenne.