Topology Optimization Considering Additive Manufacturing Process Constraints

• Author / Creator

  Shuzhi, Xu

• Additive Manufacturing (AM) is a transformative method in industrial manufacturing, facilitating the creation of lighter, stronger, and smarter parts and systems. Both plastic and metal AM technologies are extensively employed in various fields, including medical, automotive, and aeronautical industries.
  As AM provides new design opportunities, topology optimization is ideal for AM since it can be deployed to design high-performance structures and fully exploit the fabrication freedom provided by AM. However, there are still some challenges of printing topologically designed parts in AM, such as multi-material design, porous infill design, residual stress, and residual distortion, which impede its widespread use in industrial applications. These challenges can be controlled by better understanding the influence of the process or material properties used in the AM process. Nevertheless, relying exclusively on experimental efforts is expensive and time-consuming. Therefore, it is very important to consider the AM issues (such as the AM material properties, AM process model and so forth) into topology optimization.
  This research proposes a coupled topology optimization and AM process constraints system to deal with the challenges and utilize the opportunities of the AM process. Based on this system, the issues of topology optimization for AM are solved from two perspectives: improvement and prevention.
  The main objective in the perspective of ‘improvement’ is to utilize the advantages offered by AM technologies for designing high-performance parts using topology optimization algorithms. This perspective comprises two crucial contributions: one involving topology optimization methods for designing stress-based multi-material structures, and the other focusing on multi-scale porous infill structure design. In the stress-based design fabricated with multiple materials, a novel optimization algorithm that constrains the maximum stress is introduced. This new interpolation avoids the numerical issues of the extended SIMP interpolation for multi-phase stress measures. Additionally, a stress scaling method is proposed to impose the material-dependent yield stress limits. The STM based P-norm stress correction has been adopted to close the gap between the maximum local stress and the P-norm global approximation. Results of the numerical examples demonstrated that the proposed method can efficiently solve the stress constrained multi-material topology optimization problems with different material combinations. For multi-scale porous infill structure design, an innovative two-scale concurrent optimization algorithm is presented. The introduction of solid interface layers addresses the connectivity issue and thus improves the robustness of the multiscale structures. The effect of adding the interface layers has been validated through experiments and the design without interior interface layers has demonstrated an evidently degraded stiffness and strength performance. Due to the generality, the proposed methodology can reliably create optimized porous infill structures suitable for fabrication through Stereolithography (SLA) or Laser Powder Bed Fusion (LPBF).
  The perspective of ‘prevention’ is dedicated to tackling issues related to AM process defects from a preventative standpoint, specifically within the LPBF metal AM process context. These efforts primarily concentrate on mitigating defects in parts resulting from the LPBF fabrication process, such as the residual stress and distortion. Firstly, an accelerated LPBF process simulation solver is proposed based on inherent strain theory. The results are validated with a commercial software Simufact Additive 2022®, and show an average of less 5% error. Building upon this solver, a series of design methods are proposed, including a topology optimization method with constraints related to residual stress for self-supporting LPBF-fabricated parts, laser printing path optimization to minimize residual deformation in LPBF parts, support structure design that takes into account various LPBF processing constraints, and a large-scale concurrent optimization method accelerated by the PETSc framework is proposed. One featured optimization model is implemented on multiple computational cores and shows a computational advantage of almost 4.6 times over a reference case. It indicates that it is possible to reduce the residual distortion of a part by designing the part geometry, support structure, and printing path. Last but not least, the optimized structure obtained from the concurrent printing path and structure optimization has demonstrated the best structural performance (both the least residual warpage and the best stiffness).

• Subjects / Keywords

  • Topology optimization
  • additive manufacturing

• Graduation date

  Spring 2024

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-qbys-dw59

• 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.