Simulation of WAAM using TIG welding and Prediction of width and height using Machine Learning

Abstract

This thesis presents a comprehensive study on the wire arc additive manufacturing (WAAM) process using Tungsten Inert Gas (TIG) welding. The aim of the research is to investigate the thermal behavior, prediction of bead width and height, and analysis of residual stresses in WAAM structures. The study focuses on developing a better understanding of the process parameters and their influence on the final properties of the deposited layers. It investigates the residual stress distribution in WAAM structures. The stress variations from 10 to 400 MPa are observed, with stress accumulation attributed to the repetitive expansion and contraction of the locally melted region. Machine learning algorithms, including linear regression, support vector regression (SVR), and adaptive neuro-fuzzy inference system (ANFIS), are employed for the prediction of bead width and height. The models are trained using independent variables such as welding current, wire feed rate, and table speed. The accuracy of the models is evaluated using performance metrics including root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), and coefficient of determination (R-squared). In conclusion, It contributes to the understanding and optimization of the WAAM process using TIG welding. The findings provide valuable insights into the thermal behavior, prediction of bead dimensions, and analysis of residual stresses in WAAM structures.