A Quality Assessment Network for Failure Detection in 3D Printing for Future Space-Based Manufacturing

Area of Science:

• Space Manufacturing
• Additive Manufacturing
• Quality Assurance

Background:

• Space manufacturing, particularly 3D printing, is rapidly advancing with significant investment from space agencies and private companies.
• 3D printing has been successfully demonstrated in microgravity on the International Space Station (ISS), highlighting its potential for space exploration.
• Autonomous quality assessment is essential for space-based manufacturing systems operating without human intervention.

Purpose of the Study:

• To propose an automated Quality Assessment (QA) approach for space-based 3D printing.
• To enable autonomous evaluation of 3D printed parts in space, reducing reliance on human oversight.
• To develop an efficient fault detection network for common 3D printing failures in microgravity.

Main Methods:

• Investigated three common 3D printing failures: indentation, protrusion, and layering.
• Designed and implemented an automated fault detection network for quality assessment.
• Trained the network using artificial samples to evaluate its performance.

Main Results:

• The proposed QA approach achieved a detection rate of up to 82.7% for 3D printing failures.
• The system demonstrated an average confidence level of 91.6% in its assessments.
• The developed fault detection network outperformed existing network architectures.

Conclusions:

• The automated QA approach shows significant promise for reliable space-based 3D printing.
• Autonomous quality assessment is a key enabler for the future of space manufacturing.
• This technology supports the advancement of space exploration by ensuring the integrity of 3D printed components in space.