Design of Shape Forming Elements for Architected Composites via Bayesian Optimization and Genetic Algorithms: A...

Area of Science:

• Materials Science
• Additive Manufacturing
• Computational Design

Background:

• Architected composites offer tailored properties for advanced applications.
• Extrusion processes are crucial for manufacturing complex material distributions.
• Shape Forming Elements (SFEs) are novel components for multi-material extrusion.

Purpose of the Study:

• To introduce and evaluate the use of SFEs for producing architected composites via extrusion.
• To explore design automation using optimization algorithms for SFE parameter selection.
• To assess the accuracy of material distribution achieved by SFEs compared to digital models and simulations.

Main Methods:

• Development of SFEs with flow channels for multi-material routing.
• Application of Bayesian optimization and genetic algorithms for design automation.
• Minimization of objective functions: cross-section matching (pixel error + SSIM) and information content maximization (SSIM).
• Validation through extrusion of modeling clay and comparison using the Structural Similarity Index (SSIM).

Main Results:

• Rectangular flow channels in SFEs yielded better objective function values than square ones.
• Initial design-to-simulation SSIM values were ~0.8, indicating good agreement.
• Material distribution accuracy decreased significantly with successive SFE processing (SSIM dropped to 0.023).
• Flow simulations showed moderate agreement (SSIM ~0.4) but also failed to predict intended cross-sections accurately.

Conclusions:

• SFEs can impose complex material transformations during extrusion but struggle with precise material distribution control.
• Current modeling and simulation techniques lack the predictive accuracy for complex SFE-based extrusion.
• Further development of advanced modeling is required to realize the potential of SFEs in biomedical, energy storage, and structural applications.