Optimization of Wire Arc Additive Manufacturing Process Parameters for Low‐Carbon Steel and Properties Prediction by Support Vector Regression Model.

This study aims to optimize the process parameters of wire arc additive manufacturing for ER70S6 steel and to build a machine‐learning (ML) model to predict the properties of deposited specimens. Process parameters such as current, voltage, and travel speed are optimized considering other process parameters constant (gas flow rate, contact tip to the work distance, and preheat). The optimization is made using the response surface method and validated the properties by experimentation, including tensile testing and metallography. A support vector regression (SVR) ML model is implemented to predict the material's properties to substantiate the outcomes' values in every possible combination for the given parameters. In the study's findings, a significant enhancement is revealed in specimen quality, marked by reduced irregularities and porosity, and a remarkable increase in ultimate tensile strength up to 40%, validated through the SVR model. In this study, a valuable path that can be extended to predict properties of other material systems is sketched. [ABSTRACT FROM AUTHOR]