Page 86 - MSAM-1-1

Materials Science in Additive Manufacturing A ML model for AM PSP of Ti64

However, a model with all features (All) is highly for metal AM parts machining behavior prediction for
accurate to predict the AM Ti-6Al-4V machining behavior post-processing in the future.
among different AM surfaces and build orientations, due Although the S-P linkage showed excellent results in this
to the complex thermal gradients in PBF processes that study, this study was limited to only PBF AM parts. Additional
lead to different residual stress conditions and crystal studies using wrought and other AM processes, such as wire
structure. This can be attributed to the incorporation of and powder fed DED technologies, are still needed.
grain orientations information and residual stress with
SEM and machining parameters. In addition, the future Nomenclature
study will include further extending this model to similar m Local state
alloys for AM (e.g., Ti6242 - Ti-6Al-2Sn-4Zr-2Mo-0.08Si) r Vector length (μm)
and for other AM processes for Ti-6Al-4V (e.g., wire or Ψ Specimen rotation angle in XRD (°)
powder based, laser or electron beam, or plasma-based τ CRSS Critical resolved shear stress (MPa)
DED processes).
d ε Incremental strain
This study developed a new workflow to establish and θ Angular position of the cutter (°)
validate high accuracy ML models for S-P linkages based d Axial disk element thickness (mm)
on reduced-order grain morphology information for z
2
machining behavior on the PBF Ti-6Al-4V alloys. The A Angular engagement of disk element (mm )
data extraction methods were efficient and validated. In N f Number of flutes
addition, Paulson et al. (2019) have established the P-S H Helix angle (°)
linkages to connect metal AM microstructure with several R Nominal radius of the end mill (mm)
AM processing conditions based on laser power density Z Height above the free end of the cutter (mm)
and scanning strategies . Based on the findings from this T Chip thickness of the disk element (mm)
[15]
study, a full metal AM PSP linkage can be built to link AM c
processing with final post-processing machining behavior f t Feed (mm/tooth)
based on the material characterization and reduced- dF t Tangential force element (N)
ordered data science approach. dF r Radial force element (N)
K t Specific cutting energy (N/mm )
2
5. Conclusion K Ratio of radial to tangential cutting force
r
A novel Ti-6Al-4V AM workflow is presented to build an dF x Force element parallel to feed direction (N)
S-P linkage in microstructure evolution and machining dF y Force element perpendicular to feed direction (N)
behavior relationship through the utilization of advanced ε Cutting tool runout offset (mm)
data science techniques. This workflow discovered the
influence of multiple features of a grain morphology Fx Average cutting force parallel to feed direction (N)
created by the PBF process on specific cutting power Average cutting force normal to feed direction (N)
during post-AM machining. Five steps were introduced Fy
in this approach: (1) Microstructure data processing, (2)
dimensionality reduction, (3) machining data extraction, Acknowledgments
(4) extraction and evaluation of S-P linkages, and (5)
feature importance analyses. Due to the large dataset used The authors would like to acknowledge the support from
in this workflow, PCA and ML tools were developed to Dr. Saurabh Basu, Dr. Edward DeMeter, Julie Anderson,
overcome the difficulties in conventional material science Nichole Wonderling, and Fang He.
analysis. A comprehensive set of fully functional ML Conflict of interest
programs to batch process large SEM, EBSD, XRD, and
cutting force data codes were created to batch process the The authors report no conflicts of interest.
large structure and properties dataset. Author contributions

This novel workflow was highly accurate (>99%) in
predicting the machining behavior of PFB Ti-6Al-4V Conceptualization: Xi Gong, Dongrui Zeng, Guha
microstructures. Grain morphology features included Manogharan
in the workflow were microstructure spatial correlation Methodology: Xi Gong, Dongrui Zeng
functions, CLDs, residual stress, SFs, and Taylor factors,
which significantly improved the accuracy of machining Resources: Guha Manogharan
behavior prediction. This study provides a feasible routine Investigation: Xi Gong, Dongrui Zeng

Volume 1 Issue 1 (2022) 14 https://doi.org/10.18063/msam.v1i1.6