Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Discovery of Feasible 3D Printing Configurations for Metal Alloys via AI-driven Adaptive Experimental Design

Azza Fadhel, Nathaniel W. Zuckschwerdt, Aryan Deshwal, Susmita Bose, Amit Bandyopadhyay, Jana Doppa

TL;DR

This work tackles the challenge of discovering feasible additive manufacturing parameters for metal alloys under a limited experimental budget. It introduces BEAM, an AI-guided adaptive experimental design that builds a probabilistic surrogate from prior results and uses a model-informed acquisition strategy to select batches of parameter configurations for validation, balancing exploration and exploitation. The authors deploy BEAM to print NASA’s GRCop-42 using a 1000W DED system, demonstrating defect-free outputs across several laser powers within three months, and achieving substantial resource savings versus brute-force exploration. The results suggest that AI-driven adaptive experimental design can democratize access to high-performance alloys and accelerate deployment of low-power AM capabilities for aerospace applications.

Abstract

Configuring the parameters of additive manufacturing processes for metal alloys is a challenging problem due to complex relationships between input parameters (e.g., laser power, scan speed) and quality of printed outputs. The standard trial-and-error approach to find feasible parameter configurations is highly inefficient because validating each configuration is expensive in terms of resources (physical and human labor) and the configuration space is very large. This paper combines the general principles of AI-driven adaptive experimental design with domain knowledge to address the challenging problem of discovering feasible configurations. The key idea is to build a surrogate model from past experiments to intelligently select a small batch of input configurations for validation in each iteration. To demonstrate the effectiveness of this methodology, we deploy it for Directed Energy Deposition process to print GRCop--42, a high-performance copper--chromium--niobium alloy developed by NASA for aerospace applications. Within three months, our approach yielded multiple defect-free outputs across a range of laser powers dramatically reducing time to result and resource expenditure compared to several months of manual experimentation by domain scientists with no success. By enabling high-quality GRCop--42 fabrication on readily available infrared laser platforms for the first time, we democratize access to this critical alloy, paving the way for cost-effective, decentralized production for aerospace applications.

Discovery of Feasible 3D Printing Configurations for Metal Alloys via AI-driven Adaptive Experimental Design

TL;DR

This work tackles the challenge of discovering feasible additive manufacturing parameters for metal alloys under a limited experimental budget. It introduces BEAM, an AI-guided adaptive experimental design that builds a probabilistic surrogate from prior results and uses a model-informed acquisition strategy to select batches of parameter configurations for validation, balancing exploration and exploitation. The authors deploy BEAM to print NASA’s GRCop-42 using a 1000W DED system, demonstrating defect-free outputs across several laser powers within three months, and achieving substantial resource savings versus brute-force exploration. The results suggest that AI-driven adaptive experimental design can democratize access to high-performance alloys and accelerate deployment of low-power AM capabilities for aerospace applications.

Abstract

Configuring the parameters of additive manufacturing processes for metal alloys is a challenging problem due to complex relationships between input parameters (e.g., laser power, scan speed) and quality of printed outputs. The standard trial-and-error approach to find feasible parameter configurations is highly inefficient because validating each configuration is expensive in terms of resources (physical and human labor) and the configuration space is very large. This paper combines the general principles of AI-driven adaptive experimental design with domain knowledge to address the challenging problem of discovering feasible configurations. The key idea is to build a surrogate model from past experiments to intelligently select a small batch of input configurations for validation in each iteration. To demonstrate the effectiveness of this methodology, we deploy it for Directed Energy Deposition process to print GRCop--42, a high-performance copper--chromium--niobium alloy developed by NASA for aerospace applications. Within three months, our approach yielded multiple defect-free outputs across a range of laser powers dramatically reducing time to result and resource expenditure compared to several months of manual experimentation by domain scientists with no success. By enabling high-quality GRCop--42 fabrication on readily available infrared laser platforms for the first time, we democratize access to this critical alloy, paving the way for cost-effective, decentralized production for aerospace applications.
Paper Structure (18 sections, 7 equations, 4 figures, 2 tables, 1 algorithm)

This paper contains 18 sections, 7 equations, 4 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Challenges
  3. AM Process: Directed Energy Deposition
  4. General Challenges of Printing Metal Alloys
  5. Specific Challenges of Printing GRCop-42
  6. Problem Formulation
  7. Related Work
  8. AI-Guided Experimental Design to Discover Feasible Parameters
  9. Overview of BEAM Approach
  10. Key Elements of BEAM
  11. Experiments and Results
  12. Experimental Setup and Deployment Details
  13. Deployment Goals and Evaluation Metrics
  14. Results and Discussion
  15. Practical Impact and Deployment Lessons
  16. ...and 3 more sections

Figures (4)

  • Figure 1: Illustration of FormAlloy Directed energy deposition (DED)-based metal additive manufacturing (AM) system used for our deployment: (A) Build chamber showing the deposition head, powder feeding, and build plate; (B) Two powder feeders.
  • Figure 2: High-level overview of AI-guided BEAM approach to select one parameter configuration for experimental validation in each iteration (batch size $B$=1). Experiment corresponds to printing using the selected configuration followed by quality analysis to determine its success or failure.
  • Figure 3: Printed failed structures; (A) Test blocks with long column structures with the lowest being a tested structure with the columns being broken. (B) Improved test blocks with the column size increasing to much larger pillars. (C) Taller structures with the columns starting higher in the test block.
  • Figure 4: Printed successful structures from 700W laser power using DED-based metal AM; (A) As-printed cylinder. (B) Printed test block. (C) Machined cylinder with solid structure after removal of exterior shell from printing. (D) Cross-section of as-printed cylinder sample with even transition between the Inconel718 and GRCop42.