Topology Optimization for Additive Manufacturing
Author
Summary, in English
Topology optimization answers the question "How to place the material within a prescribed
design domain in order to obtain the best structural performance? " and the design
obtained is usually complex. Additive manufacturing comes with a well known design
freedom and the design provided by topology optimization can be manufactured without
as many constraints as conventional manufacturing methods. However, there does exist a
few constraints that needs to be considered such as minimum feature size, enclosed voids,
and overhang. This work focus on overcoming the overhang constraint.
A new method proposed by Langelaar (2017) solved with the optimality criteria with a
density filter provides printable structures regarding the overhang constraint. The method
is very computationally efficient, but the overhang angle is tied to the element discretization
and the printing direction needs to be axiparallel to the coordinate axis. This results
in that in order to change the inclination angle for the overhang, the element discretization
needs to be changed and the printing direction can not be optimized. Instead a new
method is proposed using the element density gradients in order alter the design to overcome
the overhang constraint. The optimization is solved using the method of moving
asymptotes with an extended density based Helmholtz PDE filter. The result shows that
the structure is affected by the added constraint. However, the provided design does not
provide completely printable structures. Further work is necessary in order to optimize
the parameters and get fully printable designs.
design domain in order to obtain the best structural performance? " and the design
obtained is usually complex. Additive manufacturing comes with a well known design
freedom and the design provided by topology optimization can be manufactured without
as many constraints as conventional manufacturing methods. However, there does exist a
few constraints that needs to be considered such as minimum feature size, enclosed voids,
and overhang. This work focus on overcoming the overhang constraint.
A new method proposed by Langelaar (2017) solved with the optimality criteria with a
density filter provides printable structures regarding the overhang constraint. The method
is very computationally efficient, but the overhang angle is tied to the element discretization
and the printing direction needs to be axiparallel to the coordinate axis. This results
in that in order to change the inclination angle for the overhang, the element discretization
needs to be changed and the printing direction can not be optimized. Instead a new
method is proposed using the element density gradients in order alter the design to overcome
the overhang constraint. The optimization is solved using the method of moving
asymptotes with an extended density based Helmholtz PDE filter. The result shows that
the structure is affected by the added constraint. However, the provided design does not
provide completely printable structures. Further work is necessary in order to optimize
the parameters and get fully printable designs.
Department/s
Publishing year
2017
Language
English
Publication/Series
ISRN LUTFD2/TFHF-17/5222-SE(1-60)
Full text
Document type
Student publication for professional degree (Master's level)
Topic
- Technology and Engineering
Keywords
- Topology optimization
- Additive manufacturing
- Overhang constraint
Supervisor
- Mathias Wallin
Scientific presentation