Abstract
The search for new alloys with improved properties is never ending with infinite combinations and amounts of alloying elements in the alloy. Advancements in machine learning have made navigating this enormous search space feasible. However, training the machine learning models and tuning their hyper-parameters to make accurate predictions can be time-consuming and often require high-performance computing resources. Furthermore, the quality of the predictions depend on the availability of sufficient training data. Here, we present a generic approach to accelerate alloy discovery by coupling high throughput CALPHAD calculations, synthetic data generation, and data mining. As a demonstration of the approach, we design super bainitic steels that form bainite at 200∘C in lower transformation times.
Original language | English |
---|---|
Article number | 115335 |
Journal | Scripta Materialia |
Volume | 228 |
DOIs | |
State | Published - Apr 15 2023 |
Funding
Notice of Copyright: This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).Research was performed at the U.S. Department of Energy's Manufacturing Demonstration Facility, located at Oak Ridge National Laboratory. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. Research was co-sponsored by the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office. Notice of Copyright: This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy . The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan ( http://energy.gov/downloads/doe-public-access-plan ).
Keywords
- Alloy design
- CALPHAD
- Data Mining
- Machine Learning
- Synthetic Data