EPJ E Highlight - Improving fluid simulations with embedded neural networks
- Details
- Published on 08 May 2023
While neural networks can help to improve the accuracy of fluid flow simulations, new research shows how their accuracy is limited unless the right approach is taken. By embedding fluid properties into neural networks, simulation accuracy can improve by orders of magnitude.
The Lattice Boltzmann Method (LBM) is a simulation technique used to describe the dynamics of fluids. Recently, there has been an increasing interest in employing neural networks for computational modelling of fluids. The results of a collaboration between researchers from Eindhoven University of Technology and Los Alamos National Laboratory, published in EPJ E, show how neural networks can be embedded into a LBM framework to model collisions between fluid particles. The team found that it is essential to embed the correct physical properties into the neural network architecture to preserve accuracy. These discoveries could deepen researchers’ understanding of how to model fluid flows.
EPJ E Highlight - Training models with a structured data curriculum
- Details
- Published on 19 April 2023
By carefully structuring the data used to train models of complex systems by leveraging physics and information theory, researchers can significantly improve the quality of their predictions, without relying on additional principles from machine learning in situations where less information about the system is available.
Researchers are now increasingly driven to identify and model the intricate mathematical patterns found in complex natural systems, where the interactions of many simple parts and subsystems can give rise to deeply intricate mathematical patterns. Today, machine learning is the most widely used technique to model these systems. Through new analysis in EPJ E, a research team at Université Paris-Saclay shows how a ‘curriculum learning’ approach, which carefully structures the data used to train models, can significantly improve their results, without relying on additional machine learning principles.
EPJ E Highlight - Shear ultrasound shaking lowers friction between solids
- Details
- Published on 15 February 2023
A simple new experiment shows how tiny ultrasound shaking at the interfaces between two objects will lower the friction between them – and in some cases, can induce sudden, large jerky motions
When high-frequency shaking occurs at an interface between two solids, recent experiments have revealed that the frictional forces between the objects can be weakened. Through a simple new experiment detailed in EPJ E, Julien Léopoldès at Université Gustave Eiffel, Marne la Vallée (formerly at ESPCI Paris) has discovered that mechanical vibrations also enhance structural aging in these systems, and can sometimes trigger sudden, jerking motions. The results could lead to a better understanding of how buildings are weakened by ambient vibrations, and may also help geologists to draw new insights into the mechanisms responsible for triggering earthquakes and landslides.
EPJ E Highlight - Machine learning could help kites and gliders to harvest wind energy
- Details
- Published on 02 February 2023
Using trial-and-error, machine learning algorithms could enable flying wind harvesters to dynamically adjust their orientations, allowing them to account for unpredictable turbulence and improve their performances.
Airborne wind energy (AWE) is a lightweight technology which uses flying devices including kites and gliders to harvest power from the atmosphere. To maximise the energy they extract, these devices need to precisely control their orientations to account for turbulence in Earth’s atmosphere. Through new research published in EPJ E, Antonio Celani and colleagues at the Abdus Salam International Center for Theoretical Physics, Italy, demonstrate how a Reinforcement Learning algorithm could significantly boost the ability of AWE devices to account for turbulence.
EPJ E Highlight - Modelling the collective movement of bacteria
- Details
- Published on 23 December 2022
Research into the movement of packages of bacteria could help better understand the formation of troublesome biofilms.
Biofilms form when microorganisms such as certain types of bacteria adhere to the surface of objects in a moist environment and begin to reproduce resulting in the excretion of a slimy glue-like substance.
These biofilms aren’t just unpleasant and unappealing however, they can be seriously troublesome. For example, in the medical field, the formation of biofilm can reduce the effectiveness of antibiotic treatments. The key to understanding biomass formation lies in understanding how bacteria behave en masse.
A new paper in EPJ E by Heinrich-Heine-Universität, Düsseldorf, Germany, researcher Davide Breoni and his co-authors presents a mathematical model for the motion of bacteria that includes cell division and death, the basic ingredients of the cell cycle.
EPJ E Topical Issue on Tissue Mechanics
- Details
- Published on 29 November 2022
Guest Editors: Alexandre Kabla, Benoît Ladoux & Jean-Marc Di Meglio
This Topical Issue of EPJ E presents a collection of contributions at the cutting edge of research in tissue mechanics.
The rich variety of subjects - epithelia submitted to different mechanical, geometrical or topological constraints, collective and cellular dynamics in cell clusters and organoids, embryology, theory of active motions mediated by topological defects, new methods of analysis – reflects the current intense activity of the biophysics community in this domain. The guest editors hope that these contributions will build a bridge between these fundamental approaches and will present the impact of physical principles on the regulation of biological tissues.
All articles are available here and are freely accessible until 28th January 2023. For further information read the Editorial.
EPJ E Topical Issue on Thermal Non-Equilibrium Phenomena in Fluid Mixtures
- Details
- Published on 12 October 2022
When a temperature difference, or gradient, is applied over a bulk fluid mixture at equilibrium, the phenomenon known as thermodiffusion, or the Ludwig-Soret effect, may occur. The thermal force will in general cause the components in the mixture to migrate until the thermal force is balanced by concentration gradients. If the thermal force is applied to a colloidal suspension, the colloids drift towards cold or hot regions. This phenomenon is commonly referred to as thermophoresis. If the fluid is soaked in a porous medium, an additional effect known as thermos-osmosis may occur. Thermo-osmosis leads to a pressure difference. These effects are different from normal diffusion and osmosis, where a concentration difference is the driving force.
EPJ E Colloquium - Thermophoresis and thermal orientation of Janus nanoparticles in thermal fields
- Details
- Published on 02 September 2022
Thermal gradients induce thermodiffusion in aqueous solutions and liquid mixtures and thermophoretic forces that drive the motion of colloids towards hot or cold regions. The Soret coefficient quantifies the strength of the thermophoretic force and varies with temperature, colloid mass and diameter, and colloid-solvent interactions. Janus colloids (JCs ) are nanoparticles with heterogeneous compositions and two contrasting properties, or "two faces" like the Roman god Janus. For example, in spherical JCs, one hemisphere might be hydrophilic and the other hydrophobic. The interest in JCs has grown steadily given their applicability in materials science. While the behaviour of JCs under equilibrium conditions has been explored, their response to thermal gradients is still not fully understood. Explaining the behaviour of JCs in a thermal field might expand their use in materials science and biomedical applications.
EPJ E Highlight - Considering how friction is maximised when liquids flow on nanoscales
- Details
- Published on 15 August 2022
By simulating a liquid confined by a nanoscale structure, researchers discovered the role molecular clogging plays in friction.
The dynamics of how liquids behave when confined in a nanoscale-sized space such as nanochannels, nanotubes or nanopores, is key to understanding a wealth of processes including lubrication, filtration and even energy storage.
The dynamics of liquids at nanoscales are different to behaviour in confinement at macroscales, however. One of the key differences that a reduction in scale creates is friction and shear between the liquid and its solid container. And further complications arise in systems with solid-to-solid contact with features like wear, micro-pitting and scuffing created.
A new paper published in EPJ E and authored by Shan Chen, from the State Key Laboratory of Organic-Inorganic Composites at Beijing University of Chemical Technology, China, uses simulations of molecular dynamics to look at the friction-induced nano-confined liquids.
EPJ E Colloquium - Predicting thermodiffusion in simple binary fluid mixtures
- Details
- Published on 14 June 2022
When a homogeneous mixture is subjected to a thermal gradient, the fluid components are partially separated because of the temperature gradient. This phenomenon, known since the mid-19th century, is called thermodiffusion, the Soret effect or thermophoresis. Despite its relatively small amplitude it impacts many natural systems, such as the salinity gradient in ocean or even pre-biological evolution, and can be exploited in applications ranging from the manipulation of biological macromolecules to isotope enrichment. However, despite numerous attempts by leading researchers, including some Nobel laureates, a full understanding of the microscopic origin of this subtle phenomenon is still lacking and there is no consensus on which model, among the numerous existing ones, is the most reliable to quantify it in dense phases.
