Multi-Material Topology Optimization Incorporating Practical Considerations in Response, Material Type Selection and Dissimilar Materials Joint

Topology Optimization (TO) is a well-established method for achieving lightweight designs. Historically, TO has evolved from considering a single candidate material to incorporating multiple candidate materials, thereby increasing design flexibility. The objective of this thesis is to propose an advanced Multi-Material Topology Optimization (MMTO) methodology emphasizing practicality in three key aspects: the integration of realistic design responses, the accommodation of hybrid isotropic and anisotropic materials, and the consideration of thin joints between dissimilar structural materials. First, an MMTO algorithm incorporating stress constraints is presented. A power-law-based stress limit interpolation scheme is proposed to ensure that different materials are assigned distinct stress limits. Additionally, a multi-material correction factor is introduced to mitigate the inherent overestimation associated with aggregation functions. To ensure numerical stability, a novel stabilization technique is developed to prevent divergence in the stress-constrained TO process. Second, an MMTO framework integrating both isotropic and anisotropic candidate materials is developed. In the proposed framework, the anisotropic material fiber orientation is treated as a concurrent design variable, eliminating the need for orientation pre-selection and enhancing solution quality. Moreover, a carry-through method is proposed to effectively calculate compliance sensitivity regarding fiber orientation without requiring strain-displacement matrix information, significantly simplifying the implementation. Third, a Multi-Joint Topology Optimization (MJTO) method is developed to consider thin joint scenarios. As a specialized branch of MMTO, MJTO considers not only multiple candidate materials but also the joint between different materials. In this method, an interface orientation coupled sandwich joint modelling method is introduced, allowing for the inclusion of thin joint in MJTO without the need for large-scale mesh refinement. Finally, the proposed methodology is evaluated using multiple models. Numerical study results indicate that the proposed stress-constrained MMTO maintains stress violations below 2%. The proposed MMTO framework integrating both isotropic and anisotropic candidate materials achieves an average performance improvement of 10% compared to the conventional method. Furthermore, the proposed MJTO method yields an average performance improvement of 20% compared to the baseline MJTO method.

Department dc:contributor.department

Mechanical and Materials Engineering

Year dc:date.issued

2026

Author dc:creator

• Shi, Yifan

Advisor dc:contributor.supervisor

• Kim, Il Yong

• Multi-Material Topology Optimization
• Multi-Joint Topology Optimization
• Lightweight Design
• Von Mises Stress
• Isotropic & Anisotropic Materials

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/1974/36252

OAI identifier oai:identifier

oai:queensu.scholaris.ca:1974/36252