Browsing by Author "Kononenko, Denys Y."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Designing materials by laser powder bed fusion with machine learning-driven bi-objective optimization
    (Rio de Janeiro : Elsevier, 2024) Kononenko, Denys Y.; Chernyavsky, Dmitry; King, Wayne E.; Hufenbach, Julia Kristin; van den Brink, Jeroen; Kosiba, Konrad
    To exploit the full industrial potential of additive manufacturing (AM) beyond prototyping, the resource-consuming identification of the optimal processing conditions needs to be minimized. This task becomes more challenging when multiple properties of the part shall be simultaneously optimized. We utilize machine learning (ML) methods in a case study on laser powder bed fusion (LPBF) of a Zr-based glass-forming alloy. Our experiments show that processing parameters affect density and amorphicity opposingly, demonstrating the efficacy of our ML-based approach. We employ multi-objective optimization using Gaussian Process Regression to model and predict target properties and their uncertainties of parts fabricated by LPBF – a widely used metal AM technology. With density and amorphicity as target parameters, we optimize models using the Pareto front facilitated by the Non-Dominated Sorting Genetic Algorithm II. Despite deviations in the amorphicity data, we demonstrate this method to identify the high-performance region of the process parameters and its ability to be iteratively enhanced with additional experimental data. This bi-objective optimization approach provides a robust toolset for navigating LPBF processing. It can be easily extended to a larger set of target properties and transferred to further AM technologies.
  • Thumbnail Image
    Item
    Machine learning for additive manufacturing: Predicting materials characteristics and their uncertainty
    (Amsterdam [u.a.] : Elsevier Science, 2023) Chernyavsky, Dmitry; Kononenko, Denys Y.; Han, Jun Hee; Kim, Hwi Jun; van den Brink, Jeroen; Kosiba, Konrad
    Additive manufacturing (AM) is known for versatile fabrication of complex parts, while also allowing the synthesis of materials with desired microstructures and resulting properties. These benefits come at a cost: process control to manufacture parts within given specifications is very challenging due to the relevance of a large number of processing parameters. Efficient predictive machine learning (ML) models trained on small datasets, can minimize this cost. They also allow to assess the quality of the dataset inclusive of uncertainty. This is important in order for additively manufactured parts to meet property specifications not only on average, but also within a given variance or uncertainty. Here, we demonstrate this strategy by developing a heteroscedastic Gaussian process (HGP) model, from a dataset based on laser powder bed fusion of a glass-forming alloy at varying processing parameters. Using amorphicity as the microstructural descriptor, we train the model on our Zr52.5Cu17.9Ni14.6Al10Ti5 (at.%) alloy dataset. The HGP model not only accurately predicts the mean value of amorphicity, but also provides the respective uncertainty. The quantification of the aleatoric and epistemic uncertainty contributions allows to assess intrinsic inaccuracies of the dataset, as well as identify underlying physical phenomena. This HGP model approach enables to systematically improve ML-driven AM processes.