A stochastic data-driven multilevel finite-element (FE2) method is introduced for random nonlinear multiscale calculations. A hybrid neural-network-interpolation (NN-I) scheme is proposed to construct a surrogate model of the macroscopic nonlinear constitutive law from representative-volume-element calculations, whose results are used as input data. Then, a FE2 method replacing the nonlinear multiscale calculations by the NN-I is developed. The NN-I scheme improved the accuracy of the neural-network surrogate model when insufficient data were available. Due to the achieved reduction in computational time, which was several orders of magnitude less than that to direct FE2, the use of such a machine-learning method is demonstrated for performing Monte Carlo simulations in nonlinear heterogeneous structures and propagating uncertainties in this context, and the identification of probabilistic models at the macroscale on some quantities of interest. Applications to nonlinear electric conduction in graphene-polymer composites are presented.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8198178 | PMC |
| http://dx.doi.org/10.3390/ma14112875 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Bottom-up Multiscale Simulation of Nonlinear Rheology for Entangled Polymer Melts: Using Atactic Polystyrene as An Example.
Macromolecules
December 2024
Pritzker School of Molecular Engineering, The University of Chicago, Chicago, Illinois 60637, United States.
Article Synopsis
- A multiscale modeling framework is developed to predict the nonlinear shear rheology of entangled polymer melts, starting from an all-atom model and transitioning to coarser models through a detailed parametrization process.
- The study focuses on atactic polystyrene melts, where the steady shear viscosity is accurately calculated across various shear rates, matching experimental data without needing adjustable parameters.
- The analysis reveals a transition from power-law behavior at high shear rates to a plateau regime at lower rates, influenced by molecular weight, allowing for comprehensive viscosity predictions over a wide range of conditions.
Multi-scale temporal and spatial variations of soil heat flux under varying riparian forests: From a day to a year.
J Environ Manage
December 2024
School of Design, Shanghai Jiao Tong University, Shanghai, 200240, China. Electronic address:
This study delves into the multi-scale temporal and spatial variations of soil heat flux (G) within riparian zones and its correlation with net radiation (Rn) across six riparian woodlands in Shanghai, each characterized by distinct vegetation types. The objective is to assess the complex interrelations between G and Rn, and how these relationships are influenced by varying vegetation and seasons. Over the course of a year, data on G and Rn is collected to investigate their dynamics.
View Article and Find Full Text PDFQuantitative electroencephalography predicts postoperative delirium in adult cardiac surgical patients from a prospective observational study.
Sci Rep
December 2024
State Key Laboratory of Bioelectronics, School of Biological Science & Medical Engineering, Southeast University, Nanjing, 210009, China.
The diagnostic and prognostic value of quantitative electroencephalogram (qEEG) in the the onset of postoperative delirium (POD) remains an area of inquiry. We aim to determine whether qEEG could assist in the diagnosis of early POD in cardiac surgery patients. We prospectively studied a cohort of cardiac surgery patients undergoing qEEG for evaluation of altered mental status.
View Article and Find Full Text PDFA spatial multiscale mathematical model of Plasmodium vivax transmission.
J Math Biol
December 2024
School of Mathematics and Statistics, The University of Melbourne, Parkville, Australia.
The epidemiological behavior of Plasmodium vivax malaria occurs across spatial scales including within-host, population, and metapopulation levels. On the within-host scale, P. vivax sporozoites inoculated in a host may form latent hypnozoites, the activation of which drives secondary infections and accounts for a large proportion of P.
View Article and Find Full Text PDFODD-Net: a hybrid deep learning architecture for image dehazing.
Sci Rep
December 2024
Department of Mechatronics, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India.
Haze can significantly reduce visibility and contrast of images captured outdoors, necessitating the enhancement of images. This degradation in image quality can adversely affect various applications, including autonomous driving, object detection, and surveillance, where poor visibility may lead to navigation errors and obscure crucial details. Existing dehazing techniques face several challenges: spatial methods tend to be computationally heavy, transform methods often fall short in quality, hybrid methods can be intricate and demanding, and deep learning methods require extensive datasets and computational power.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!