Laser Powder Bed Fusion (LPBF) is increasingly being used to create smart materials such as Shape Memory Alloys (SMAs) in aerospace, automotive, and medical applications. The additive manufacture of these materials allows for the creation of novel geometries that would be impossible to create using standard methods. The prediction of Transformation Temperatures (TTs) is a critical design feature of SMA systems that defines the activation threshold of the two unique characteristics: shape memory and superelasticity effects. Because of the numerous variables at play, utilizing experimental methods to investigate and anticipate the impact of LPBF process parameters on TTs would be costly and time-consuming. Machine learning (ML) approaches have the potential to drastically reduce the number of experimental attempts required to optimize the alloy design process.
This talk will examine alternative ways for defining a practical AM processing routine for SMAs using ML, as well as offer an insight on future achievements in this domain.
- Describe the effect of the AM processing parameters on the final shape memory alloy material properties.
- Define the role of machine learning in accelerating material-AM relationships optimization.