Filters

2020 - 20232

More filters

2021 - 20232
Reset filters

Settings

End date: 2023 × Start date: 2020 × DDC: 530 × DDC: 690 × Type: Text ×

Search Results

Now showing 1 - 2 of 2
  • 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.
  • Thumbnail Image
    Item
    The impact of atmospheric boundary layer, opening configuration and presence of animals on the ventilation of a cattle barn
    (Amsterdam [u.a.] : Elsevier Science, 2020) Nosek, Štěpán; Kluková, Zuzana; Jakubcová, Michaela; Yi, Qianying; Janke, David; Demeyer, Peter; Jaňour, Zbyněk
    Naturally ventilated livestock buildings (NVLB) represent one of the most significant sources of ammonia emissions. However, even the dispersion of passive gas in an NVLB is still not well understood. In this paper, we present a detailed investigation of passive pollutant dispersion in a model of a cattle barn using the wind tunnel experiment method. We simulated the pollution of the barn by a ground-level planar source. We used the time-resolved particle image velocimetry (TR-PIV) and the fast flame ionisation detector (FFID) to study the flow and dispersion processes at high spatial and temporal resolution. We employed the Proper Orthogonal Decomposition (POD) and Oscillating Patterns Decomposition (OPD) methods to detect the coherent structures of the flow. The results show that the type of atmospheric boundary layer (ABL) and sidewall opening height have a significant impact on the pollutant dispersion in the barn, while the presence of animals and doors openings are insignificant under conditions of winds perpendicular to the sidewall openings. We found that the dynamic coherent structures, developed by the Kelvin-Helmholtz instability, contribute to the pollutant transport in the barn. We demonstrate that in any of the studied cases the pollutant was not well mixed within the barn and that a significant underestimation (up to by a factor 3) of the barn ventilation might be obtained using, e.g. tracer gas method. © 2020 The Authors