Feature-based Level Set Topology Optimization and Its Multidisciplinary Applications

• Author / Creator

  Liu, Jikai

• When parts are designed and manufactured, considerations for their geometry, material, function, and manufacturability need to be taken into account, and it would be efficient to address all these aspects concurrently. From this perspective, topology optimization is useful and gaining the popularity, especially for complex and multidisciplinary design problems which can be difficult to solve by experience-based methods. Extensive research efforts have been reported on enhancing the problem solving capability of topology optimization, but its relationship with feature-based design is rarely explored which leads to a gap between different stages of design activities. Therefore, the generic research problem is the limited involvement of feature technology in topology optimization. To be specific, a tedious post-treatment is mandatory to transform the freeform topological design into a geometry feature-based CAD model to support the following feature-based design activities. This issue hurdles the effective application of topology optimization and has a negative impact on the overall design efficiency and quality. In addition, the feature attached semantic information is rarely considered by the topology optimization algorithms. Because of this, some practical measures such as manufacturing time and cost cannot be optimally designed. To fix the problems, the generic research theme is to realize the feature-based level set topology optimization. The research activities are divided into two stages: develop geometry feature-based topology optimization approaches, and then gradually involve the feature-attached semantic information in the optimization algorithms. Specifically, essence of these research activities is presented in the following paragraphs. First, several geometry feature-based topology optimization approaches have been developed targeting at different types of design problems. To be specific, the implicit feature-based approach realizes the predefined geometry pattern in the topological design, and an application example is the constant rib thickness design of the injection molding parts. Compared to the existing methods, this implicit feature-based approach is simpler and more efficient in numerical implementation. The explicit feature based approach realizes the geometry feature based design in the explicit form, and an application example is the 2.5D machining part design. According to the literature survey result, the explicit feature insertion capability is a novel contribution. Then, the B-rep geometry-based approach facilitates the topology optimization implementation in CAD systems and this is of great engineering significance from the integration perspective. Second, the feature-attached semantic information is gradually involved to further enhance the geometry feature-based topology optimization approaches. Taking a machining feature for example, the attached semantic information can facilitate the quantitative evaluations of the part’s manufacturing processes, tooling, machining time, costs, etc. If these evaluations could be analytically expressed based on the level set function, they can be configured into the topology optimization problem under the level set framework. Then, through the proper solution, both the part’s functionality and manufacturability can be designed. Apparently, the multi-stage design requirements are concurrently addressed during the early conceptual design, which definitely shortens the product development process and improves the product’s profitability and competitiveness. On this basis, a new scheme - optimization-for-manufacture (OFM), is proposed, which is an extension of the conventional design-for-manufacture (DFM) for better material use efficiency and design quality for manufacture. For the research outcome, they are summarized into six chapters. All these contributions are built based on a common methodology of level set topology optimization, and they are presented in a progressive order from pure geometry feature-based approaches to the semantic information involvement. At the end, the novel industrial applications are demonstrated.

• Subjects / Keywords

  • Level set
  • Topology optimization
  • Feature-based design

• Graduation date

  Spring 2016

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3TD9NP72

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.