Predicting Structural Responses of Manufactured Models via 3D Printing through Machine Learning

This study investigates the application of machine learning to predict structural responses of auxetic models manufactured via 3D printing. Using algorithms such as random forest, neural networks and polynomial regression, predictive models were developed to estimate the Poisson parameters, mass and relative frequency of the parts before manufacturing. The methodology was divided into two main stages: experimental data collection, where detailed information about the structural properties of the parts was collected, and training and validation of machine learning models, ensuring the accuracy of the predictions. The results demonstrate that the models provide accurate predictions, contributing significantly to the optimization of the additive manufacturing process and ensuring the structural integrity of the printed auxetic parts. This work highlights the potential of machine learning in materials engineering and manufacturing, especially in the context of complex structures such as auxetic.