AbstractThe key issue restricting the use of computer-aided design (CAD) models within an optimization framework is that there is no clear definition of how the change in CAD parameters effect the model’s performance in terms of optimizing for certain objective functions (for e.g. minimum pressure loss, minimum drag, maximum lift etc.). In this thesis, an automated optimization process is presented, which uses the parameters defining the features in a feature-based CAD model as design variables. This process exploits adjoint methods for the computation of gradients, which predicts how the objective function changes for an infinitesimally small movement of each surface mesh node in the normal direction. The use of adjoint methods results in a computational cost that is essentially independent of the number of design variables, making it ideal for optimization in a large parameter space.
The success of any shape optimization methodology depends on the choice of parameters and can sometimes stifle the creation of high performing innovative solutions. Parametric effectiveness is a measure that rates the ability of the parameters in a model to change its shape in the optimum way. Here, the optimum shape change is that suggested by the adjoint sensitivity on the model boundary. Herein, an automated approach is developed to compute the parametric effectiveness of CAD model parameters. In cases where the parametric effectiveness is low, a novel methodology is shown which automatically adds the optimum features to the CAD model feature tree, and thus increases the design freedom of the model. In this thesis, the optimization framework is developed to exploit the capabilities of modern CAD systems to add geometrical constraints to the optimization process including minimum thickness, constant volume and packaging constraints. The packaging constraints are imposed by the adjacent components in the CAD model product assembly which the component being optimized is not allowed to violate.
The applicability of the developed approaches is demonstrated on a range of CAD models created in CATIA V5 for 2D and 3D finite element and computational fluid dynamics problems. During this research, the ability to carry out optimization directly on the CAD models created in commercial CAD systems has been enhanced. In addition, it has been shown that the additional shape flexibility imparted to the model by inserting additional “optimum” CAD features, leads to a better optimized component than would have been possible using the original model. Lastly, it has been shown that an optimization process can be configured to respect CAD assembly constraints, resulting in an optimized geometry that does not violate the space occupied by other components in the product assembly.
|Date of Award||29 Jun 2018|
|Supervisor||Trevor Robinson (Supervisor) & Cecil Armstrong (Supervisor)|