Macroscale Property Prediction for Additively Manufactured IN625 from Microstructure through Advanced Homogenization

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Modeling

Background:

• Additive manufacturing (AM) of metallic parts necessitates accurate computational models.
• Predicting mechanical behavior requires integrating microstructural, manufacturing, and operational factors.

Purpose of the Study:

• To predict the mechanical response of IN625 tensile coupons under varying conditions for the AFRL AM Modeling Challenge 3.
• To develop and evaluate advanced computational methods for AM part simulation.

Main Methods:

• A representative volume element (RVE) approach combined with crystal plasticity within the Fast Fourier Transformation (FFT) framework.
• Introduction of proper generalized decomposition (PGD) for advanced material model identification.
• Exploration of Self-consistent Clustering Analysis (SCA) as an alternative to FFT.

Main Results:

• Successful application of RVE and FFT-based crystal plasticity for predicting IN625 mechanical response.
• Demonstration of PGD as an effective method for material model calibration challenges.
• Presentation of SCA as a viable alternative reduced-order modeling technique.

Conclusions:

• The developed computational framework accurately predicts the mechanical behavior of AM IN625.
• Advanced material identification and reduced-order modeling techniques enhance simulation capabilities for AM.
• The study provides insights into modeling assumptions and physical interpretations for AM parts.