Additive Structure Topology Optimization Without Support Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated design solutions for additive manufacturing lack an integrated manufacturing process that uses sensitivity-based optimization to eliminate the need for support structures, leading to increased production costs and time due to the requirement for support structures in printing overhangs less than a defined angle.
Innovation Solution
A method employing penalty functions in topology optimization to modify design variables, eliminating or reducing the need for support structures by applying a penalty function to design variables prior to filtering, which transforms original mathematical values into physical design variables suitable for additive manufacturing, ensuring that overhangs greater than a threshold angle are printed without support structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If support structures are used for printing overhangs less than a defined angle, then manufacturing reliability is improved, but manufacturing cost and production time increase
Solution Approach 1:
The penalty function is applied during the topology optimization phase before manufacturing to preemptively eliminate design configurations that would require support structures. By modifying design variables with the penalty function f(θ) = cos(θ)/cos(α) where θ is the overhang angle and α is the critical angle, the optimization process automatically adjusts the structure to avoid steep overhangs, ensuring manufacturability without support structures before the actual printing begins.
2Reliability
If support structures are used for printing overhangs less than a defined angle, then manufacturing reliability is improved, but material usage increases
Solution Approach 1:
The penalty function modifies the objective function in topology optimization to penalize designs requiring support structures. The function f(θ) = cos(θ)/cos(α) creates a gradient that guides the optimization toward designs with gentler overhangs, eliminating the need for support material entirely and reducing overall material consumption.
3Ease of manufacture
If penalty functions are applied to design variables in topology optimization, then the need for support structures is eliminated, but optimization complexity increases
Solution Approach 1:
The penalty function transforms the design variable ρ (material density) by applying a multiplicative factor based on the overhang angle θ: ρ_physical = ρ_mathematical × f(θ), where f(θ) = cos(θ)/cos(α). This parameter transformation integrates manufacturing constraints directly into the optimization variables, automatically guiding the design toward manufacturable configurations without requiring complex constraint algorithms.
4Adaptability or versatility
If designs are optimized without considering manufacturing constraints, then design freedom is maximized, but manufacturing cost increases
Solution Approach 1:
The penalty function seamlessly integrates manufacturing constraints into the topology optimization framework by modifying the physical design variables. The transformation ρ_physical = ρ_mathematical × cos(θ)/cos(α) allows the optimization to explore a wide design space while automatically filtering out configurations that would require support structures, maintaining design freedom while ensuring cost-effective manufacturability.
Data Source
AI summary
One goal in automated product designing of additive manufacturing is to obtain designs having overhangs without support structures if the criterion for overhangs is rigorously geometrical. In an embodiment of the present invention, designers can request automated optimization and design, using simulation and sensitivity-based optimization, of structures having overhangs in the print direction that do not need any support structures. In an embodiment, a method includes, at a processor, calculating model design responses and model sensitivities of a computer-aided engineering (CAE) model in a CAE system based on design variables of the CAE model for various design responses being either applied in objective or constraints. The method further includes optimizing values of the design variables. The method further includes calculating physical design variables by employing a penalty function. Additionally, the calculations can also be in conjunction with employing material interpolation schemes. The method further includes generating an optimized CAE model using the calculated physical design variables. The optimized CAE model is free or partly free of support structures. The method further includes printing the optimized CAE model being free or partly free of support structures.