OPTIMIZATION OF COMPUTATIONAL TIME FOR DIGITAL TWIN DATABASE IN DIRECTED ENERGY DEPOSITION FOR RESIDUAL STRESSES

Date

2023

Authors

Tariq, Usman
Joy, Ranjit
Wu, Sung-Heng
Arif Mahmood, Muhammad
Woodworth, Michael M.
Liou, Frank

Journal Title

Journal ISSN

Volume Title

Publisher

University of Texas at Austin

Abstract

Metal Additive Manufacturing (MAM) has experienced rapid growth and demonstrated its cost-effectiveness in the production of high-quality products. However, MAM processes introduce significant thermal gradients that result in the formation of residual stresses and distortions in the final parts. Finite Element Analysis (FEA) is a valuable tool for predicting residual stresses, but it requires substantial computational power. This study aims to reduce computational time by incorporating a thermo-mechanical model specifically designed for the Directed Energy Deposition (DED) process using Ti6Al4V. This model predicts the thermal history and subsequent residual stresses in the deposited material. Various FEA methods, including “chunk”, layer, and conventional methods are examined, providing a comparative analysis of computational cost and numerical accuracy. These findings contribute towards the realization of a digital twin database, where the incorporation of efficient and accurate FEA models can optimize part quality and strength while reducing computational time.

Description

LCSH Subject Headings

Citation