About this Abstract |
| Meeting |
2027 TMS Annual Meeting & Exhibition
|
| Symposium
|
AI-Enabled Materials Processing: Integrating Accelerated Experimental Workflows and Processing-Aware Machine Learning
|
| Presentation Title |
AlloyBot SmartMelting: Closed-Loop Process Control for Reliable Automatic Arc-Melting Synthesis |
| Author(s) |
Vignesh Selvaraj, Ming En Pek, Sebastian A. Kube |
| On-Site Speaker (Planned) |
Vignesh Selvaraj |
| Abstract Scope |
Through automation and innovative design, our AlloyBot arc-melting system has enabled the synthesis of over 150 bulk alloys per week. As next development stage, we present closed-loop process control to produce high-quality alloys even for challenging compositions: Various process parameters such as the arc-current, duration, and number of remelt cycles determine the quality of arc-melted alloys, captured by metrics such as mixing uniformity, porosity, impurity pickup, and compositional accuracy. In conventional arc-melting, skilled operators guide the melting process through intuition and visual feedback. Here, as integral part of AlloyBot automatic synthesis, we replace operator intuition with intelligent closed-loop process control. In-situ sensing provides real-time feedback during melting, enabling on-the-fly adjustment of process parameters. Ex-situ characterization feeds a learning loop that predicts optimal process parameters a priori for new compositions. Through case studies across different alloy systems, we demonstrate how alloy quality can be systematically targeted and optimized. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Process Technology, Computational Materials Science & Engineering, Machine Learning |