| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Novel Strategies for Rapid Acquisition and Processing of Large Datasets from Advanced Characterization Techniques
|
| Presentation Title |
Overview of Machine Learning in Low-Latency Automated Data Analysis for In-Situ Synchrotron X-ray Diffraction in Metals and Alloys |
| Author(s) |
Tingkun Liu, Vinay Amatya, Venkata B Vukkum, Arun Devaraj |
| On-Site Speaker (Planned) |
Tingkun Liu |
| Abstract Scope |
Probing microstructural evolution under realistic processing environments using in-situ synchrotron x-ray diffraction (SXRD) is crucial for understanding material behaviors and optimizing microstructure in metals and alloys. The large volume of low latency data generated by SXRD experiments poses challenges for efficient analysis and interpretation. Machine learning techniques can be leveraged to accelerate in-situ SXRD data analysis and extract meaningful insights about microstructural evolution. We are aiming to develop an automated capability for real-time in operando experimental analysis of non-equilibrium microstructural evolution during laser-based additive manufacturing of complex alloys. This will be achieved by integrating in operando experimental approaches with machine learning models for data processing and analysis, which will be used by a software defined architecture framework to achieve low power, low latency processing parameter control. Here, we provide an overview of machine learning methodologies in low-latency automated data analysis for in-situ SXRD in metals and alloys. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Characterization, Machine Learning |