| About this Abstract |
| Meeting |
Materials Science & Technology 2019
|
| Symposium
|
Advanced Manufacturing, Processing, Characterization and Modeling of Functional Materials
|
| Presentation Title |
P2-45: Machine Learning for the 3D Printing of Polymer Composite Inks |
| Author(s) |
Isidro Pantoja G, Irmak Sargin, Sepehr Nesaei, Clarence C King, Arda Gozen, Scott P Beckman |
| On-Site Speaker (Planned) |
Isidro Pantoja G |
| Abstract Scope |
The goal of this project is to quantitatively predict the structure and properties of 3D printed components from polymer composite inks based on the bulk physical properties of their constituents and printing parameters. Here we explore the relationship between the physical properties of ink constituents and the viscoelastic behavior of the ink. In particular, we examine inks comprised of two different molecular weights of polyethylene oxide with varying loads of graphene. The inks are characterized via extensional and shear viscosity measurements. A fair amount of data has been collected and cleansed. The classification has begun, while data collection continues. A classification study has been carried out to identify the extensional relaxation modes based on the compositions. Promising results with an accuracy of around 80 percent have been obtained. Simple regression methods have been used to develop the basis for later more complex analysis of machine learning. |