Computers & Structures Journal
An Interface-Preserving Level Set Update Strategy for Topology Optimization of Mechanical Assemblies
Structural components in assemblies often require specific geometric features — such as cylindrical regions for joints — to function correctly. Standard topology optimization methods, however, struggle to impose geometric, feature-preserving constraints on selected boundary regions during shape updates. We propose a shape update strategy for level set-based topology optimization of mechanical assemblies that enables constrained shape updates along user-specified boundaries while allowing free-form updates elsewhere. The constrained regions are limited to affine motions such as translation, rotation, and scaling, providing greater control that is especially valuable in engineering design. This is particularly useful for multi-functional components in larger assemblies, where certain boundaries must retain primitive geometries and vary only within specified limits. For example, when a component must contain a cylindrical aperture to fit a pin of unknown radius, our method allows simultaneous optimization of the aperture’s location, orientation, and size, alongside the component’s overall topology. We extend the standard Hilbert space extension method by introducing its constrained variant which incorporates affine motion constraints into the velocity extension. The resulting velocity field satisfies descent direction requirements for the optimization while ensuring that all feature-preserving constraints are met. We demonstrate the method’s effectiveness on canonical structural problems with geometrically constrained boundaries.
Download publicationAssociated Researchers
Related Publications
2025
RECALL-MM: A Multimodal Dataset of Consumer Product Recalls for Risk Analysis using Computational Methods and Large Language ModelsNew multi-modal design dataset contains historical information about…
2024
Evaluating Large Language Models for Material SelectionThis work evaluates the use of LLMs for predicting materials of…
2024
Optimal Design of Vehicle Dynamics Using Gradient-Based, Mixed-Fidelity Multidisciplinary OptimizationThis research showcases a multidisciplinary approach to optimize a…
2024
A hyperreduced reduced basis element method for reduced-order modeling of component-based nonlinear systemsThis method balances accuracy and computational speed through adaptive…
2024
Elicitron: An LLM Agent-Based Simulation Framework for Design Requirements ElicitationA novel framework that leverages Large Language Models (LLMs) to…
Get in touch
Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.
Contact us