Abstract
Recent hybrid additive-subtractive manufacturing platforms can usher a materials-on-demand manufacturing paradigm where the materials, along with geometry and surface morphology, are designed to meet the functionality. While in situ measurement systems have advanced to monitor geometric and morphological variations, microstructure and phase characterization methods remain slow, impeding this vision. Towards addressing this issue, we present an approach to characterize the material concurrently during fabrication by associating in situ sensor measurements with the microstructures. Analysis of sensor data from machining of 316L steel suggests that the microstructural variations under different process parameters can be predicted in real time with 93% accuracy.
Original language | English (US) |
---|---|
Pages (from-to) | 29-33 |
Number of pages | 5 |
Journal | Manufacturing Letters |
Volume | 23 |
DOIs | |
State | Published - Jan 2020 |
Externally published | Yes |
Keywords
- High-resolution sensing
- Hybrid manufacturing
- Machine learning
- Machining dynamics
- Microstructure characterization
ASJC Scopus subject areas
- Mechanics of Materials
- Industrial and Manufacturing Engineering