FYFD

Tag: computational fluid dynamics

  • Featured Video Play Icon

    Exascale Simulations

    Capturing what goes on inside a combustion engine is incredibly difficult. It’s a problem that depends on turbulent flow, chemistry, heat transfer, and more. To represent all of those aspects in a numerical simulation requires enormous computational resources. It’s not simply the realm of a supercomputer; it requires some of the fastest supercomputers in existence.

    Exascale computing, like that used for the simulations in this video, is defined as at least 10^18 (floating-point) operations per second. For comparison, my PC has a recent, high-end graphics card, and it’s about a million times slower than that. These are absolutely gigantic simulations. (Image and video credit: N. Wimer et al.)

  • Why Moths Are Slow Fliers

    Why Moths Are Slow Fliers

    Hawkmoths and other insects are slow fliers compared to birds, even ones that can hover. To understand why these insects top out at 5 m/s, researchers simulated their flight from hovering to forward flight at 4 m/s. They analyzed real hawkmoths flying in wind tunnels to build their simulated insects, then studied their digital moths with computational fluid dynamics.

    During hovering flight, they found that hawkmoths generate equal amounts of lift with their upstroke and downstroke. As the moth transitions into forward flight, though, its wing orientation shifts to reduce drag, and the upstroke stops being so helpful. Instead, the upstroke generates a downward lift that the downstroke has to counter in addition to the insect’s weight. At higher forward speeds, this trend gets even worse.

    The final verdict? Hawkmoths don’t have the flexibility to twist their wings on the upstroke the way birds do to avoid that large downward lift. Since they can’t mitigate that negative lift, the insects have a slower top speed overall. (Image and research credit: S. Lionetti et al.; via APS Physics; submitted by Kam-Yung Soh)

  • Escaping the Sun

    Escaping the Sun

    One enduring mystery of the solar wind — a stream of high-energy particles expelled from the sun — is how the particles get accelerated in the first place. The sun frequently belches out spurts of plasma, but without further momentum, that material simply falls back to the sun’s surface under the star’s gravity. Mechanisms like shock waves can further accelerate particles that are already moving quickly, but they cannot explain how the particles get going in the first place.

    A recent study used supercomputers to tackle this challenging problem in turbulent plasma physics. Each simulation tracked nearly 200 billion particles, requiring tens of thousands of processors. The results showed that turbulence itself provides the necessary initial acceleration and serves as the first step to getting particles moving fast enough to escape the sun. (Image credit: NASA SDO; research credit: L. Comisso and L. Sironi; via Physics World)

  • Inside a Champagne Pop

    Inside a Champagne Pop

    When the cork pops on a bottle of champagne, the physics is akin to that of a missile launch in more ways than one. In this study, researchers used computational fluid dynamics to closely examine the gases that escape behind the cork. They identified three phases to the flow. In the first, the exhaust gases form a crown-shaped expansion region, complete with shock diamonds. Once the cork has moved far enough downstream, the axial flow accelerates to supersonic speeds and a bow shock forms behind the cork. Finally, the pressure in the bottle drops low enough that supersonic conditions cannot be maintained and the flow becomes subsonic. (Image credit: top – Kindel Media, simulation – A. Benidar et al.; research credit: A. Benidar et al.; via Ars Technica; submitted by Kam-Yung Soh)

    A numerical simulation showing the ejection of a champagne cork from a bottle. The colors indicate the speed of gases escaping from the bottle.
    A numerical simulation showing the ejection of a champagne cork from a bottle. The colors indicate the speed of gases escaping from the bottle.
  • Asperitas Formation

    Asperitas Formation

    In 2017, the World Meteorological Organization named a new cloud type: the wave-like asperitas cloud. How these rare and distinctive clouds form is still a matter of debate, but this new study suggests that they need conditions similar to those that produce mammatus clouds, plus some added shear.

    Using direct numerical simulations, the authors studied a moisture-filled cloud layer sitting above drier ambient air. Without shear, large droplets in this cloud layer slowly settle downward. As the droplets evaporate, they cool the area just below the cloud, changing the density and creating a Rayleigh-Taylor-like instability. This is one proposed mechanism for mammatus clouds, which have bulbous shapes that sink down from the cloud.

    When they added shear to the simulation, the authors found that instead of mammatus clouds, they observed asperitas ones. But the amount of shear had to be just right. Too little shear produced mammatus clouds; too much and the shear smeared out the sinking lobes before they could form asperitas waves. (Image credit: A. Beatson; research credit: S. Ravichandran and R. Govindarajan)

  • Re-Entry For X-Wings

    Re-Entry For X-Wings

    Fans of sci-fi and fantasy have a long-standing tradition of exploring the physics and/or practicality of creations in their fandom, and Star Wars fans are no exception. Here engineers ask whether Luke Skywalker’s X-wing fighter could survive the descent through Dagobah’s atmosphere as he searched for Master Yoda. Their results are based on a numerical simulation, with some assumptions about the spacecraft’s descent path and design as well as the planet’s atmosphere. Fans of the Jedi will be glad to hear that the X-wing can survive its supersonic descent intact, delivering the last Jedi safely to his mentor. (Image credit: Y. Ling et al.)

  • Featured Video Play Icon

    Burning Virtual Forests

    Wildfires are growing ever more frequent and more destructive as the climate crisis worsens. Unfortunately, simulating and predicting the course of these fires is incredibly difficult, requiring a combination of ecology, meteorology, combustion science, and more. To handle so many variables, model builders often turn to statistics that allow them to simulate an entire forest but at the cost of representing individual trees as a few pixels or a cone.

    In this video, researchers show a new wildfire simulation based on a computationally efficient but more realistic depiction of trees. With individual, three-dimensional trees, the simulation can capture effects that are otherwise hard to examine – like the difference in burn rate for coniferous and deciduous forests and the likelihood that a fire can jump a firebreak of a given size. Their weather, fire, and atmospheric models are even able to simulate the birth of fire-generated clouds! Check out the full video to see more and then head over to their site if you’d like to dig into the methodology. (Video and research credit: T. Hädrich et al.; see also)

  • Mountains in the Sky

    Mountains in the Sky

    Our skies can sometimes presage the weather to come. In thunderstorms, a cirrus plume above an anvil cloud will often appear (visible by satellite) about half an hour before severe conditions are reported on the ground. A new study delves into the origins of these plumes and finds that they result from an internal hydraulic jump in the storm that acts a bit like an artificial mountain, driving air — and the moisture it contains — higher in the stratosphere than normal. Once the jump is established, the authors found it could drive 7 tonnes per second of water vapor into the stratosphere! (Image credit: jplenio; research credit: M. O’Neill et al.; via Science)

  • Marshland Wave Damping

    Marshland Wave Damping

    Coastal marshes are a critical natural defense against flooding. The flexible plants of the marsh both slow the water’s current and help damp waves. As a result of that hydrodynamic dissipation, marshes help protect against erosion and reduce the magnitude of flooding events. But coastal managers looking to maintain or improve their marshes in order to mitigate climate-change-driven storms need to be able to predict what level of vegetation they need.

    To that end, a team of researchers has built a new model to better capture the flow effects of marsh grasses. Building from an individual, flexible plant (as opposed to a rigid cylinder, as grass is often represented), the authors constructed a model able to predict wave dissipation for many marsh configurations, which should help better predict the infrastructure changes needed in different coastal regions. (Image credit: T. Marquis; research credit: X. Zhang and H. Nepf; via APS Physics)

  • Superfluid Instabilities

    Superfluid Instabilities

    Superfluids — like Bose-Einstein condensates — are bizarre compared to fluids from our everyday experience because they have no viscosity. Without viscosity, it’s no surprise that they behave in unusual ways. Here, researchers simulated superfluids moving past one another. In each of these images, the blue fluid is moving to the left, and the red fluid is moving to the right. In a typical fluid, such motion causes ocean-wave-like curls due to the Kelvin-Helmholtz instability.

    Instead, with a low relative velocity and high repulsion between atoms of the two layers, the superfluids form a tilted, finger-like interface (Image 1) that the authors refer to as a flutter-finger pattern. (Repulsion essentially sets the miscibility between the superfluids. With a high repulsion, the superfluids resist mixing.)

    With a higher relative velocity (Image 2), the wavelength of the ripples becomes comparable to the thickness of the interface, and the superfluids take on a very different, zipper-like pattern. Note how the tips detach and reconnect to the neighboring finger!

    With lower repulsion, the interface between the two liquids is thicker and breaks down quickly (Image 3). The authors call this a sealskin pattern. (Image credits: water – M. Blažević, simulations – H. Kokubo et al.; research credit: H. Kokubo et al.; via APS Physics)