| Abstract Scope |
Additive manufacturing (AM), particularly fused filament fabrication (FFF), is defined by the range of available materials and modifiable variables. This research focuses on developing a Machine Learning algorithm to predict surface quality, internal defects, and geometric accuracy by analyzing FFF surgical guides using Nylon 12 and ULTEM 1010, separately. Parts will be fabricated with varying print parameters, including raster angle, print orientation, infill pattern, infill density, and body width. The resultant parts will be evaluated for surface quality using a laser confocal microscope. The image data will be analyzed to establish correlations between printing variables and average surface roughness (Ra), as well as provide comparisons between ULTEM 1010 and Nylon 12. This study aims to enhance the understanding of printing parameter effects of surface quality on novel FFF-produced parts in a practical biomedical application as well as to develop a versatile ML algorithm applicable to other materials and industries. |