arXiv subjects

Alex Skillen

Publications and source records attributed to Alex Skillen.

7 recordsLinked to original sources

arXiv2026

Scale by scale analysis of magnetoconvection with uniform wall-normal and wall-parallel magnetic fields at low magnetic Reynolds number

Rayleigh-B\'enard convection under an imposed inductionless magnetic field is analysed statistically from the perspective of single-point and multi-scale energy budgets. The data is obtained from direct numerical simulations with a Rayleigh number of $10^6$, a Prandtl number of $1$ and Hartmann numbers of $0$, $20$, $40$ and $80$. Wall-parallel and wall-normal magnetic fields are considered as two separate cases. The initial analysis focuses qualitatively on the influence of the magnetic field upon the coherent structures. A central contribution of this work is the interpretation of these structural modifications through magnetohydrodynamically modified turbulent kinetic energy budgets. For example, in the wall-normal case, the thinning of the thermal plumes can be attributed to the damping of the pressure-diffusion mechanisms due to the Lorentz dissipation. In the wall-parallel configuration, Joule dissipation induces a pressure-strain redistribution mechanism that preferentially transfers kinetic energy from the wall-normal velocity component to the field-perpendicular, wall-parallel velocity component but less so to the field-parallel velocity component. This description is then extended to scale-space by considering budgets relating second- and third-order structure functions. Here, the anisotropy is accounted for by analysing directional structure functions. Despite the anisotropy, the Lorentz force appears as an isotropic sink damping intermediate and large scales of motion. The result of this is a lack of transfer between scales of motion and hence a flow with suppressed small-scale turbulence. These results establish a link between qualitative observations and long-term energy balances, providing new insight into magnetoconvective turbulence and informing future modelling and theoretical approaches to such flows.

physics.flu-dyn↗

arXiv2026

Unstable magnetic reconnection self-generates turbulence

Magnetic reconnection and turbulence are deeply intertwined in magnetohydrodynamic flows, yet how reconnection self-generates turbulence remains unclear. Using an ensemble of high-resolution three-dimensional direct numerical simulations of an unstable magnetised jet with an initially weak mean field, we demonstrate a self-sustained transition from laminar reconnection to fully developed turbulence in the absence of external forcing. We show that a three-dimensional current-sheet instability triggers stochastic reconnection, leading to persistent turbulent energy injection. Energy-budget analysis reveals that the coupling between the turbulent electromotive force and the magnetic mean shear dominates turbulent production, with magnetic fluctuations subsequently transferring energy to the kinetic field through a nonlinear cascade.

physics.plasm-ph↗

arXiv2025

Learning Temporally Consistent Turbulence Between Sparse Snapshots via Diffusion Models

We investigate the statistical accuracy of temporally interpolated spatiotemporal flow sequences between sparse, decorrelated snapshots of turbulent flow fields using conditional Denoising Diffusion Probabilistic Models (DDPMs). The developed method is presented as a proof-of-concept generative surrogate for reconstructing coherent turbulent dynamics between sparse snapshots, demonstrated on a 2D Kolmogorov Flow, and a 3D Kelvin-Helmholtz Instability (KHI). We analyse the generated flow sequences through the lens of statistical turbulence, examining the time-averaged turbulent kinetic energy spectra over generated sequences, and temporal decay of turbulent structures. For the non-stationary Kelvin-Helmholtz Instability, we assess the ability of the proposed method to capture evolving flow statistics across the most strongly time-varying flow regime. We additionally examine instantaneous fields and physically motivated metrics at key stages of the KHI flow evolution.

physics.flu-dyn↗

arXiv2024

Concerning the Use of Turbulent Flow Data for Machine Learning

This article describes some common issues encountered in the use of Direct Numerical Simulation (DNS) turbulent flow data for machine learning. We focus on two specific issues; 1) the requirements for a fair validation set, and 2) the pitfalls in downsampling DNS data before training. We attempt to shed light on the impact these issues can have on machine learning and computer vision for turbulence. Further, we include statistical and spectral analysis for the homogenous isotropic turbulence from the John Hopkins Turbulence Database, a Kolmogorov flow, and a Rayleigh-B\'enard Convection Cell using data generated by the authors, to concretely demonstrate these issues.

physics.flu-dyn↗

arXiv2023

Spectrally Decomposed Diffusion Models for Generative Turbulence Recovery

We investigate the statistical recovery of missing physics and turbulent phenomena in fluid flows using generative machine learning. Here we develop a two-stage super-resolution method using spectral filtering to restore the high-wavenumber components of a Kolmogorov flow. We include a rigorous examination of generated samples through the lens of statistical turbulence. By extending the prior methods to a combined super-resolution and conditional high-wavenumber generation, we demonstrate turbulence recovery on a 8x upsampling task, effectively doubling the range of recovered wavenumbers.

physics.flu-dyn↗

arXiv2023

The Propulsion Performance of Dual Heaving Foils in Tandem, with Spanwise Flexibility

The impact of spanwise flexibility on the propulsion performance of two foils arranged in tandem and subjected to a prescribed sinusoidal heaving motion has been studied at a Reynolds number of 100. This comprises a wide range of natural frequencies of the flexible foil, covering pre-resonance, resonance, and post-resonance regimes, with respect to the heaving frequency. The propulsion performance of the foils under different Strouhal numbers, with either fixed reduced frequency or fixed natural frequency, has also been investigated. We observed that the propulsion efficiency, thrust, and power consumption drop, but the tip deformation coefficient increases to its maximum when operating at resonance. The propulsion efficiency of the flexible dual foils increases significantly in the pre-resonance regime at small Strouhal numbers. Both thrust force and power consumption increase with the reduced frequency during the pre- and post-resonance regimes. An instability phenomenon is observed in the instantaneous tip deformation coefficient of the flexible dual foils in the post-resonance regimes. We also find that the dual heaving foils, with a larger resonance Strouhal number, achieve a greater thrust increment compared with that of rigid dual heaving foils when working at the pre-resonance regimes. Meanwhile, dual heaving foils with a smaller resonance Strouhal number can achieve a larger propulsion efficiency increment compared with that of rigid dual heaving foils when working at small Strouhal numbers.

physics.flu-dyn↗

arXiv2019

Thermal transients in a U-Bend

We study numerically the propagation of a hot thermal transient through a U-bend via an ensemble of wall-resolved large eddy simulations. Conjugate heat transfer between fluid and solid domains is accounted for. The flow is in a fully turbulent mixed convection regime, with a bulk Reynolds number of $10,000$, a Richardson number of $2.23$, and water as the working fluid (Prandtl number = $6$). These conditions lead to strong thermal stratification, with buoyancy-induced secondary flows, and the generation of a large and persistent recirculation region. The evolution of Dean vortices as the thermal transient passes is studied. It is found that baroclinic vorticity generation dominates over a large period of the transient, due to the thermal inertia of the wall. Gravitational buoyancy leads to a reversal of the counter-rotating vortex pair. The impact of this reversal on the swirl-switching and secondary-current losses is assessed. It is found that low frequency modes are suppressed in the reversed-vortex state.

physics.flu-dyn↗