CAD-Based Parameterization for Adjoint Optimization
Manipulating CAD geometry using primitive components rather than the originating software is typically a challenging prospect. The parameterisation used to define the geometry of a model is often integral to the efficiency of the design. Even more crucial are the relations (constraints) between those parameters that do not allow the model to be under-defined. However, access to these parameters is lost when making the CAD model portable. Importing a standard CAD file gives access to the Boundary Representation (BRep) of the model and consequently its boundary surfaces which are usually trimmed patches. Therefore, in order to connect Adjoint optimization and Computational Fluid Dynamics to the industrial design framework (CAD) in a generic manner, the BRep must be used as a starting point to produce volume meshes and as a means of changing a model’s shape. In this study, emphasis is given firstly, to meshing (triangulation) of a BRep model as a precursor to volume meshing and secondly, to the use of techniques similar to Free Form Deformation for changing the model’s shape.
The work shown here is part of the IODA (Industrial Optimal Design using Adjoint CFD) Project. Research topic: Intuitive interfaces for optimisation parameterisation, constraint definition and automated mesh-to-CAD conversion.
The project leading to this application, has received funding from the European Unions Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No. 642959.
- 4.Nowacki, H., Dannenberg, L.: Approximation Methods Used in the Exchange of Geometric Information via the VDA/VDMA Surface Interface. Product Data Interfaces in CAD/CAM Applications, 150–159 (1986)Google Scholar
- 6.Mantyla, M. (1988). An introduction to solid modeling. Rockville, MD: Computer Science Press.Google Scholar
- 8.TUM Department of Mechanical Engineering, Chair of Aerodynamics and Fluid mechanics. Retrieved from: www.aer.mw.tum.de/en/research-groups/automotive/drivaer
- 13.Meagher, D.: Geometric modeling using octree encoding. Comput. Vision Graph, 19(2), 129–147 (1982)Google Scholar
- 14.Isaac, T., Burstedde, C., Ghattas, O. (2012) : Low-Cost Parallel Algorithms for 2:1 Octree Balance. 2012 IEEE 26th International Parallel and Distributed Processing SymposiumGoogle Scholar
- 17.Tristano, J. R., Owen, S. J., Canann, S. A.: Advancing Front Surface Mesh Generation in Parametric Space Using a Riemannian Surface Definition (1998), https://doi.org/10.1.1.36.630
- 18.Herrmann, L. R.: Laplacian-isoparametric grid generation scheme. J. Eng. Mech. Div-ASCE, 102(5), 749–756 (1976)Google Scholar
- 26.Giles, M. B., Duta, M. C., Mller, J.-D., and Pierce, N. A.: Algorithm developments for discrete adjoint methods, (2001).Google Scholar
- 30.Othmer, C.: Implementation of a continuous adjoint for topology optimization of ducted flows, 2007 (2007).Google Scholar
- 34.Yan, X., Su, X. G.: Linear Regression Analysis (2009), https://doi.org/10.1142/6986