Related Experiment Video
Updated: Sep 19, 2025
08:32
Indirect Fabrication of Lattice Metals with Thin Sections Using Centrifugal Casting
Published on: May 14, 2016
12.6K
High-throughput alloy and process design for metal additive manufacturing.
Sofia Sheikh1, Brent Vela1, Pejman Honarmandi1
1Department of Materials Science and Engineering, Texas A&M University, College Station, TX USA.
Npj Computational Materials
|June 16, 2025
Summary
This study developed a computational framework to predict alloy printability for additive manufacturing (AM). It uses deep learning to rapidly assess defects, enabling efficient novel alloy design.
Area of Science:
- Materials Science
- Computational Engineering
- Additive Manufacturing
Background:
- Conventional engineering alloys often lack suitability for additive manufacturing (AM).
- Assessing alloy printability requires complex analysis of composition and processing parameters.
- Experimental evaluation is time-consuming, necessitating high-throughput computational methods.
Purpose of the Study:
- To introduce a computational framework for assessing alloy printability in AM.
- To accelerate the evaluation of process-induced defects like lack-of-fusion, balling, and keyholing.
- To facilitate the design of novel alloys optimized for AM.
Main Methods:
- Integration of material properties, processing parameters, and thermal models.
- Utilizing three thermal models to predict melt pool profiles.
- Development of a deep learning surrogate model for accelerated printability assessment.
- Validation using printability maps for the CoCrFeMnNi system.
Main Results:
- The framework accurately assesses printability and predicts defects.
- A deep learning model speeds up assessment by 1000x without accuracy loss.
- Printability maps were generated for the CoCrFeMnNi system.
- Exploration of printable alloys within the high-entropy alloy space was achieved.
Conclusions:
- The developed framework efficiently navigates large alloy design spaces for AM.
- Probabilistic printability maps offer insights into defect likelihood and uncertainty.
- This approach enhances the design of new alloys for additive manufacturing.
