Evaluating Process Parameters Impact On Deformation In 3D Printing: A Machine Learning And Simulation Study

Abstract

This paper presents a deformation prediction analysis for 3D printed parts, employing machine learning models alongside finite element simulations to develop a comprehensive dataset. We utilized five machine learning algorithms—Linear Regression, Decision Tree Regressor, Random Forest Regressor, Support Vector Regressor, and K-Nearest Neighbors Regressor—to assess the influence of critical process parameters, including layer height, print temperature, print speed, and bed temperature, on the deformation of printed components. Finite element simulations were conducted to generate accurate deformation data, which served as the foundation for training the machine learning models. Each algorithm's performance was rigorously evaluated, revealing insights into their predictive capabilities and the significant effects of the analyzed parameters on deformation outcomes. Importantly, the study also contributes to environmental protection and sustainable manufacturing by reducing material waste, minimizing failed prints, and lowering energy consumption, thereby supporting environmental management and green technologies. The results provide valuable guidance for optimizing 3D printing processes, ultimately enhancing printed parts' mechanical performance, reliability, and sustainability